Welcome to privacyIDEA

privacyIDEA is a modular authentication system. Using privacyIDEA you can enhance your existing applications like local login, VPN, remote access, SSH connections, access to web sites or web portals with a second factor during authentication. Thus boosting the security of your existing applications. Originally it was used for OTP authentication devices. But other “devices” like challenge response and SSH keys are also available. It runs on Linux and is completely Open Source, licensed under the AGPLv3.

privacyIDEA can read users from many different sources like flat files, different LDAP services, SQL databases and SCIM services. (see Realms)

Authentication devices to provide two factor authentication can be assigned to those users, either by administrators or by the users themselves. Policies define what a user is allowed to do in the web UI and what an administrator is allowed to do in the management interface.

The system is written in python, uses flask as web framework and an SQL database as datastore. Thus it can be enrolled quite easily providing a lean installation. (see Installation)

Table of Contents

Overview

privacyIDEA is a system that is used to manage devices for two factor authentication. Using privacyIDEA you can enhance your existing applications like local login, VPN, remote access, SSH connections, access to web sites or web portals with a second factor during authentication. Thus boosting the security of your existing applications.

In the beginning there were OTP tokens, but other means to authenticate like SSH keys are added. Other concepts like handling of machines or enrolling certificates are coming up, you may monitor this development on Github.

privacyIDEA is a web application written in Python based on the flask micro framework. You can use any webserver with a wsgi interface to run privacyIDEA. E.g. this can be Apache, Nginx or even werkzeug.

A device or item used to authenticate is still called a “token”. All token information is stored in an SQL database, while you may choose, which database you want to use. privacyIDEA uses SQLAlchemy to map the database to internal objects. Thus you may choose to run privacyIDEA with SQLite, MySQL, PostgreSQL, Oracle, DB2 or other database.

The code is divided into three layers, the API, the library and the database layer. Read about it at Code Documentation. privacyIDEA provides a clean REST API.

Administrators can use a Web UI or a command line client to manage authentication devices. Users can log in to the Web UI to manage their own tokens.

Authentication is performed via the API or certain plugins for FreeRADIUS, simpleSAMLphp, Wordpress, Contao, Dokuwiki... to either provide default protocols like RADIUS or SAML or to integrate into applications directly.

Due to this flexibility there are also many different ways to install and setup privacyIDEA. We will take a look at common ways to setup privacyIDEA in the section Installation but there are still many others.

Installation

The ways described here to install privacyIDEA are

• the installation via the Python Package Index, which can be used on any Linux distribution and
• ready made Ubuntu Packages for Ubuntu 14.04LTS and
• ready made Debian Packages for Debian Wheezy.

If you want to upgrade from a privacyIDEA 1.5 installation please read Upgrading.

privacyIDEA needs python 2.7 to run properly!

Python Package Index

You can install privacyidea on usually any Linux distribution in a python virtual environment. This way you keep all privacyIDEA code in one defined subdirectory.

Note

privacyIDEA depends on python 2.7 to run properly.

You first need to install some development packages. E.g. on debian based distributions the packages are called

• libjpeg-dev
• libz-dev
• python-dev
• libffi-dev
• libssl-dev
• libxslt1-dev

Now you can install privacyIDEA like this:

virtualenv /opt/privacyidea

cd /opt/privacyidea
source bin/activate

Now you are within the python virtual environment. Within the environment you can now run:

pip install --upgrade setuptools
pip install privacyidea

Please see the section The Config File for a quick setup of your configuration.

Then create the encryption key and the signing keys:

pi-manage create_enckey
pi-manage create_audit_keys

Create the database and the first administrator:

pi-manage createdb
pi-manage admin add admin -e admin@localhost

Now you can run the server for your first test:

pi-manage runserver

Depending on the database you want to use, you may have to install additional packages.

Footnotes

[1]	https://launchpad.net/~privacyidea

[2]	https://github.com/privacyidea/privacyidea/tree/master/authmodules/simpleSAMLphp

Ubuntu Packages

There are ready made packages for Ubuntu 14.04 LTS and 16.04 LTS [4]. These are available in a public ppa repository [1], so that the installation will automatically resolve all dependencies. Install it like this:

add-apt-repository ppa:privacyidea/privacyidea
apt-get update

There are the base packages python-privacyidea and the administrator tool privacyideaadm.

But we recommend installing the meta package:

apt-get install privacyidea-apache2

which will install the code, the webserver and the database and configure everything accordingly. If you do not like the Apache2 webserver you could alternatively use the meta package privacyidea-nginx.

After installing with Apache2 or Nginx you only need to create your first administrator and you are done:

pi-manage admin add admin -e admin@localhost

Now you may proceed to First Steps.

Note

The packages privacyidea-apache2 and privacyidea-nginx assume that you want to run a privacyIDEA system. These packages deactivate all other (default) websites. You can install the package privacyidea-mysql to install the privacyIDEA application and setup the database. After this, you need to configure the webserver on your own.

Note

To get the latest development snapshots, you can use the repository ppa:privacyidea/privacyidea-dev. But these packages might be broken sometimes!

FreeRADIUS

privacyIDEA has a perl module to “translate” RADIUS requests to the API of the privacyIDEA server. This module plugs into FreeRADIUS. The FreeRADIUS does not have to run on the same machine like privacyIDEA. To install this module run:

apt-get install privacyidea-radius

For further details see RADIUS plugin.

SimpleSAMLphp

Starting with 1.4 privacyIDEA also supports SAML via a plugin for simpleSAMLphp [2]. The simpleSAMLphp service does not need to run on the same machine like the privacyIDEA server.

To install it on a Ubuntu 14.04 system please run:

apt-get install privacyidea-simplesamlphp

For further details see simpleSAMLphp Plugin.

PAM

privacyIDEA also comes with a PAM library to add two factor authentication to any Linux system. You can run one central privacyIDEA server and configure all other systems using the PAM library to authenticate against this privacyIDEA.

To install it on a Ubuntu 14.04 system please run:

apt-get install privacyidea-pam

For further details see Pluggable Authentication Module.

OTRS

OTRS is an important Open Source Ticket Request System. It is written in Perl and privacyIDEA provides an authentication plugin to authenticate at OTRS with two factors.

To install it on Ubuntu 14.04 please run:

apt-get install privacyidea-otrs

For further details and configuration see OTRS.

Footnotes

[1]	https://launchpad.net/~privacyidea

[2]	https://github.com/privacyidea/privacyidea/tree/master/authmodules/simpleSAMLphp

[3]	http://www.otrs.com/

[4]	Starting with privacyIDEA 2.15 Ubuntu 16.04 packages are provided

Debian Packages

Wheezy

You can install privacyIDEA on Debian Wheezy either via the Python Package Index or with a ready made Wheezy package.

The available Wheezy package privacyidea-venv_2.1~dev0_amd64.deb contains a complete virtual environment with all necessary dependent modules. To install it run:

dpkg -i privacyidea-venv_2.1~dev0_amd64.deb

This will install privacyIDEA into a virtual environment at /opt/privacyidea/privacyidea-venv.

You can enter the virtual environment by:

source /opt/privacyidea/privacyidea-venv/bin/activate

Jessie

At the moment you can use the Ubuntu Trusty packages with Debian Jessie.

Thus you can create a file /etc/apt/sources.list.d/privacyidea.list with the content:

deb http://ppa.launchpad.net/privacyidea/privacyidea/ubuntu trusty main

Add the GPG key to the keyring:

gpg --keyserver keyserver.ubuntu.com --recv-keys C24DCF7D
gpg --armor --export C24DCF7D | apt-key add -

Now run:

apt-get update
apt-get install privacyidea-apache2

As an alternative you can find a complete guideline how to setup privacyIDEA including RADIUS here [3].

Running privacyIDEA with Apache2 and MySQL

If you installed via pip or the Wheezy package you need to create and fill the config directory /etc/privacyidea manually:

cp /opt/privacyidea/privacyidea-venv/etc/privacyidea/dictionary \
/etc/privacyidea/

Create a config /etc/privacyidea/pi.cfg like this:

# Your database
SQLALCHEMY_DATABASE_URI = 'mysql://pi:password@localhost/pi'
# This is used to encrypt the auth_token
SECRET_KEY = 'choose one'
# This is used to encrypt the admin passwords
PI_PEPPER = "choose one"
# This is used to encrypt the token data and token passwords
PI_ENCFILE = '/etc/privacyidea/enckey'
# This is used to sign the audit log
PI_AUDIT_KEY_PRIVATE = '/etc/privacyidea/private.pem'
PI_AUDIT_KEY_PUBLIC = '/etc/privacyidea/public.pem'
PI_LOGFILE = '/var/log/privacyidea/privacyidea.log'
#CRITICAL = 50
#ERROR = 40
#WARNING = 30
#INFO = 20
#DEBUG = 10
PI_LOGLEVEL = 20

You need to create the above mentioned logging directory /var/log/privacyidea.

You need to create the above mentioned database with the corresponding user access:

mysql -u root -p -e "create database pi"
mysql -u root -p -e "grant all privileges on pi.* to 'pi'@'localhost' \
identified by 'password'"

With this config file in place you can create the database tables, the encryption key and the audit keys:

pi-manage createdb
pi-manage create_enckey
pi-manage create_audit_keys

Now you can create the first administrator:

pi-manage admin add administrator

The system is set up. You now only need to configure the Apache2 webserver.

The Apache2 needs a wsgi script that could be located at /etc/privacyidea/piapp.wsgi and look like this:

import sys
sys.stdout = sys.stderr
from privacyidea.app import create_app
# Now we can select the config file:
application = create_app(config_name="production", \
config_file="/etc/privacyidea/pi.cfg")

Finally you need to create a Apache2 configuration /etc/apache2/sites-available/privacyidea.conf which might look like this:

WSGIPythonHome /opt/privacyidea/privacyidea-venv
<VirtualHost _default_:443>
     ServerAdmin webmaster@localhost
     # You might want to change this
     ServerName localhost

     DocumentRoot /var/www
     <Directory />
             # For Apache 2.4 you need to set this:
             # Require all granted
             Options FollowSymLinks
             AllowOverride None
     </Directory>

     # We can run several instances on different paths with different configurations
     WSGIScriptAlias /      /etc/privacyidea/piapp.wsgi
     #
     # The daemon is running as user 'privacyidea'
     # This user should have access to the encKey database encryption file
     WSGIDaemonProcess privacyidea processes=1 threads=15 display-name=%{GROUP} user=privacyidea
     WSGIProcessGroup privacyidea
     WSGIPassAuthorization On

     ErrorLog /var/log/apache2/error.log

     LogLevel warn
     LogFormat "%h %l %u %t %>s \"%m %U %H\"  %b \"%{Referer}i\" \"%{User-agent}i\"" privacyIDEA
     CustomLog /var/log/apache2/ssl_access.log privacyIDEA

     #   SSL Engine Switch:
     #   Enable/Disable SSL for this virtual host.
     SSLEngine on

     #   If both key and certificate are stored in the same file, only the
     #   SSLCertificateFile directive is needed.
     SSLCertificateFile    /etc/ssl/certs/privacyideaserver.pem
     SSLCertificateKeyFile /etc/ssl/private/privacyideaserver.key

     <FilesMatch "\.(cgi|shtml|phtml|php)$">
             SSLOptions +StdEnvVars
     </FilesMatch>
     <Directory /usr/lib/cgi-bin>
             SSLOptions +StdEnvVars
     </Directory>
     BrowserMatch ".*MSIE.*" \
             nokeepalive ssl-unclean-shutdown \
             downgrade-1.0 force-response-1.0

</VirtualHost>

The configuration assumes, a user privacyidea, which you need to create:

useradd -r -m privacyidea

The files in /etc/privacyidea and the logfiles in /var/log/privacyidea/ should be restricted to this user.

Footnotes

[1]	https://launchpad.net/~privacyidea

[2]	https://github.com/privacyidea/privacyidea/tree/master/authmodules/simpleSAMLphp

[3]	http://www.routerperformance.net/howtos/install-privacyidea-2-13-on-a-clean-debian-8-jessie/

CentOS Installation

There is a detailed Howto [1] for installing privacyIDEA with FreeRADIUS 3 on CentOS 7 using a python virtual environment.

RPM Repository

For customers with a valid service level agreement [2] with NetKnights there is an RPM repository, that can be used to easily install and update privacyIDEA on CentOS 7 / RHEL 7. For more information see [3].

[1]	https://www.privacyidea.org/two-factor-authentication-with-otp-on-centos-7/

[2]	https://netknights.it/en/leistungen/service-level-agreements/

[3]	https://netknights.it/en/additional-service-privacyidea-support-customers-centos-7-repository/

Upgrading

If you installed privacyIDEA via DEB or RPM repository you can use the normal system ways of apt-get, aptitude and rpm to upgrade privacyIDEA to the current version.

Basic pip upgrade process

If you install privacyIDEA into a python virtualenv like /opt/privacyidea, you can follow this basic upgrade process.

First you might want to backup your program directory:

tar -zcf privacyidea-old.tgz /opt/privacyidea

and your database:

source /opt/privacyidea/bin/activate
pi-manage backup create

Running upgrade

Starting with version 2.17 the script privacyidea-pip-update performs the update of the python virtualenv and the DB schema.

Just enter your python virtualenv (you already did so, when running the backup) and run the command:

privacyidea-pip-update

Manual upgrade

Now you can upgrade the installation:

source /opt/privacyidea/bin/activate
pip install --upgrade privacyidea

Usually you will need to upgrade/migrate the database:

pi-manage db stamp 4f32a4e1bf33 -d /opt/privacyidea/lib/privacyidea/migrations
pi-manage db upgrade -d /opt/privacyidea/lib/privacyidea/migrations

Now you need to restart your webserver for the new code to take effect.

Upgrade to privacyIDEA 2.12

In privacyIDEA 2.12 the Event Handler framework was added. Two new tables “eventhandler” and “eventhandleroption” were added.

You need to update the database models:

pi-manage db stamp 4f32a4e1bf33 -d path/to/migrations
pi-manage db upgrade -d path/to/migrations

Upgrade to privacyIDEA 2.11

In privacyIDEA 2.11 the RADIUS server definition was added. RADIUS servers can be used in RADIUS tokens and in the RADIUS passthru policy.

A new database table “radiusserver” was added.

You need to update the database models:

pi-manage db stamp 4f32a4e1bf33 -d path/to/migrations
pi-manage db upgrade -d path/to/migrations

Upgrade to privacyIDEA 2.10

In privacyIDEA 2.10 SMTP servers were added. SMTP servers can be used for notifications, registration and also for Email token and SMS token.

SMTP servers need a new database table “smtpserver”.

You need to update the database models:

pi-manage db stamp 4f32a4e1bf33 -d path/to/migrations
pi-manage db upgrade -d path/to/migrations

privacyIDEA 2.10 can import all kind of PSKC token files. These XML files need to be parsed. Therefore BeautifulSoup4 and lxml is used. On pip installations you need to install a package like libxslt1-dev.

Upgrade From privacyIDEA 2.x to 2.3

In 2.3 the priority of resolvers in realms was added.

You need to update the database models:

pi-manage db stamp 4f32a4e1bf33 -d path/to/migrations
pi-manage db upgrade -d path/to/migrations

Note

You need to specify the path to the migrations scripts. This could be /usr/lib/privacyidea/migrations.

Note

When upgrading with the Ubuntu LTS packages, the database update is performed automatically.

Upgrade From privacyIDEA 1.5

Warning

privacyIDEA 2.0 introduces many changes in database schema, so at least perform a database backup!

Stopping Your Server

Be sure to stop your privacyIDEA server.

Upgrade Software

To upgrade the code enter your python virtualenv and run:

pip install --upgrade privacyidea

Configuration

Read about the configuration in the The Config File.

You can use the old enckey, the old signing keys and the old database uri. The values can be found in your old ini-file as privacyideaSecretFile, privacyideaAudit.key.private, privacyideaAudit.key.public and sqlalchemy.url. Your new config file might look like this:

config_path = "/home/cornelius/tmp/pi20/etc/privacyidea/"
# This is your old database URI
# Note the three slashes!
SQLALCHEMY_DATABASE_URI = "sqlite:///" + config_path + "token.sqlite"
# This is new!
SECRET_KEY = 't0p s3cr3t'
# This is new
#This is used to encrypt the admin passwords
PI_PEPPER = "Never know..."
# This is used to encrypt the token data and token passwords
# This is your old encryption key!
PI_ENCFILE = config_path + 'enckey'
# THese are your old signing keys
# This is used to sign the audit log
PI_AUDIT_KEY_PRIVATE = config_path + 'private.pem'
PI_AUDIT_KEY_PUBLIC = config_path + 'public.pem'

To verify the new configuration run:

pi-manage create_enckey

It should say, that the enckey already exists!

Migrate The Database

You need to upgrade the database to the new database schema:

pi-manage db upgrade -d lib/privacyidea/migrations

Note

In the Ubuntu package the migrations folder is located at /usr/lib/privacyidea/migrations/.

Create An Administrator

With privacyIDEA 2.0 the administrators are stored in the database. The password of the administrator is salted and also peppered, to avoid having a database administrator slip in a rogue password.

You need to create new administrator accounts:

pi-manage addadmin <email-address> <admin-name>

Start The Server

Run the server:

pi-manage runserver

or add it to your Apache or Nginx configuration.

The Config File

privacyIDEA reads its configuration from different locations:

1. default configuration from the module privacyidea/config.py

2. then from the config file /etc/privacyidea/pi.cfg if it exists and then

3. from the file specified in the environment variable PRIVACYIDEA_CONFIGFILE.

   export PRIVACYIDEA_CONFIGFILE=/your/config/file

The configuration is overwritten and extended in each step. I.e. values define in privacyidea/config.py that are not redefined in one of the other config files, stay the same.

You can create a new config file (either /etc/privacyidea/pi.cfg) or any other file at any location and set the environment variable. The file should contain the following contents:

# The realm, where users are allowed to login as administrators
SUPERUSER_REALM = ['super', 'administrators']
# Your database
SQLALCHEMY_DATABASE_URI = 'sqlite:////etc/privacyidea/data.sqlite'
# This is used to encrypt the auth_token
SECRET_KEY = 't0p s3cr3t'
# This is used to encrypt the admin passwords
PI_PEPPER = "Never know..."
# This is used to encrypt the token data and token passwords
PI_ENCFILE = '/etc/privacyidea/enckey'
# This is used to sign the audit log
PI_AUDIT_KEY_PRIVATE = '/home/cornelius/src/privacyidea/private.pem'
PI_AUDIT_KEY_PUBLIC = '/home/cornelius/src/privacyidea/public.pem'
# PI_AUDIT_MODUL = <python audit module>
# PI_AUDIT_SQL_URI = <special audit log DB uri>
# PI_LOGFILE = '....'
# PI_LOGLEVEL = 20
# PI_INIT_CHECK_HOOK = 'your.module.function'
# PI_CSS = '/location/of/theme.css'
# PI_UI_DEACTIVATED = True

Note

The config file is parsed as python code, so you can use variables to set the path and you need to take care for indentations.

SQLALCHEMY_DATABASE_URI defines the location of your database. You may want to use the MySQL database or Maria DB. There are two possible drivers, to connect to this database. Please read MySQL database connect string.

The SUPERUSER_REALM is a list of realms, in which the users get the role of an administrator.

PI_INIT_CHECK_HOOK is a function in an external module, that will be called as decorator to token/init and token/assign. This function takes the request and action (either “init” or “assing”) as an arguments and can modify the request or raise an exception to avoid the request being handled.

There are three config entries, that can be used to define the logging. These are PI_LOGLEVEL, PI_LOGFILE, PI_LOGCONFIG. These are described in Debugging and Logging.

You can use PI_CSS to define the location of another cascading style sheet to customize the look and fell. Read more at Themes.

Note

If you ever need passwords being logged in the log file, you may set PI_LOGLEVEL = 9, which is a lower log level than logging.DEBUG. Use this setting with caution and always delete the logfiles!

privacyIDEA digitally signs the responses. You can disable this using the parameter PI_NO_RESPONSE_SIGN. Set this to True to suppress the response signature.

You can set PI_UI_DEACTIVATED = True to deactivate the privacyIDEA UI. This can be interesting if you are only using the command line client or your own UI and you do not want to present the UI to the user or the outside world.

Note

The API calls are all still accessable, i.e. privacyIDEA is technically fully functional.

Audit parameters

PI_AUDIT_MODULE lets you specify an alternative auditing module. The default which is shipped with privacyIDEA is privacyidea.lib.auditmodules.sqlaudit. There is no need to change this, unless you know exactly what you are doing.

You can change the servername of the privacyIDEA node, which will be logged to the audit log using the variable PI_AUDIT_SERVERNAME.

You can run the database for the audit module on another database or even server. For this you can specify the database URI via PI_AUDIT_SQL_URI.

PI_AUDIT_TRUNCATE = True lets you truncate audit entries, that to the length of the database fields.

In certain cases when you experiencing problems you may use the parameters PI_AUDIT_POOL_SIZE and PI_AUDIT_POOL_RECYCLE.

Debugging and Logging

You can set PI_LOGLEVEL to a value 10 (Debug), 20 (Info), 30 (Warning), 40 (Error) or 50 (Critical). If you experience problems, set PI_LOGLEVEL = 10 restart the web service and resume the operation. The log file privacyidea.log should contain some clues.

You can define the location of the logfile using the key PI_LOGFILE. Usually it is set to:

PI_LOGFILE = "/var/log/privacyidea/privacyidea.log"

Advanced Logging

You can also define a more detailed logging by specifying a log configuration file like this:

PI_LOGCONFIG = "/etc/privacyidea/logging.cfg"

Such a configuration could look like this:

[formatters]
keys=detail

[handlers]
keys=file,mail

[formatter_detail]
class=privacyidea.lib.log.SecureFormatter
format=[%(asctime)s][%(process)d][%(thread)d][%(levelname)s][%(name)s:%(lineno)d] %(message)s

[handler_mail]
class=logging.handlers.SMTPHandler
level=ERROR
formatter=detail
args=('mail.example.com', 'privacyidea@example.com', ['admin1@example.com',\
   'admin2@example.com'], 'PI Error')

[handler_file]
# Rollover the logfile at midnight
class=logging.handlers.RotatingFileHandler
backupCount=14
maxBytes=10000000
formatter=detail
level=DEBUG
args=('/var/log/privacyidea/privacyidea.log',)

[loggers]
keys=root,privacyidea

[logger_privacyidea]
handlers=file,mail
qualname=privacyidea
level=DEBUG

[logger_root]
level=ERROR
handlers=file

The file structure follows [1] and can be used to define additional handlers like logging errors to email addresses.

Note

In this example a mail handler is defined, that will send emails to certain email addresses, if an ERROR occurs.

Footnotes

[1]	https://docs.python.org/2/library/logging.config.html#configuration-file-format

The WSGI Script

Apache2 and Nginx are using a WSGI script to start the application.

This script is usually located at /etc/privacyidea/privacyideaapp.py or /etc/privacyidea/privacyideaapp.wsgi and has the following contents:

import sys
sys.stdout = sys.stderr
from privacyidea.app import create_app
# Now we can select the config file:
application = create_app(config_name="production",
                         config_file="/etc/privacyidea/pi.cfg")

In the create_app-call you can also select another config file.

Note

This way you can run several instances of privacyIDEA in one Apache2 server by defining several WSGIScriptAlias definitions pointing to different wsgi-scripts, that again reference different config files with different database definitions.

Running Apache instances

To run further Apache instances add additional lines in your Apache config:

WSGIScriptAlias /instance1 /etc/privacyidea1/privacyideaapp.wsgi
WSGIScriptAlias /instance2 /etc/privacyidea2/privacyideaapp.wsgi
WSGIScriptAlias /instance3 /etc/privacyidea3/privacyideaapp.wsgi
WSGIScriptAlias /instance4 /etc/privacyidea4/privacyideaapp.wsgi

It is a good idea to create a subdirectory in /etc for each instance. Each wsgi script needs to point to the corresponding config file pi.cfg.

Each config file can define its own

• database
• encryption key
• signing key
• ...

To create the new database you need the command pi-manage. The command pi-manage reads the configuration from /etc/privacyidea/pi.cfg.

If you want to use another instance with another config file, you need to set an environment variable and create the database like this:

PRIVACYIDEA_CONFIGFILE=/etc/privacyidea3/pi.cfg pi-manage createdb

This way you can use pi-manage for each instance.

The pi-manage Script

pi-manage is the script that is used during the installation process to setup the database and do many other tasks.

Note

The interesting thing about pi-manage is, that it does not need the server to run as it acts directly on the database. Therefor you need read access to /etc/privacyidea/pi.cfg and the encryption key.

If you want to use a config file other than /etc/privacyidea/pi.cfg, you can set an environment variable:

PRIVACYIDEA_CONFIGFILE=/home/user/pi.cfg pi-manage

pi-manage always takes a command and sometimes a sub command:

pi-manage <command> [<subcommand>] [<parameters>]

For a complete list of commands and sub commands use the -h parameter.

You can do the following tasks.

Encryption Key

You can create an encryption key and encrypt the encryption key.

Create encryption key:

pi-manage create_enckey

Note

This command takes no parameters. The filename of the encryption key is read from the configuration. The key will not be created, if it already exists.

The encryption key is a plain file on your hard drive. You need to take care, to set the correct access rights.

You can also encrypt the encryption key with a passphrase. To do this do:

pi-manage encrypt_enckey /etc/privacyidea/enckey

and pipe the encrypted enckey to a new file.

Read more about the database encryption and the enckey in Security Modules.

Backup and Restore

You can create a backup which will be save to /var/lib/privacyidea/backup/.

The backup will contain the database dump and the complete directory /etc/privacyidea. You may choose if you want to add the encryption key to the backup or not.

Warning

If the backup includes the database dump and the encryption key all seeds of the OTP tokens can be read from the backup.

As the backup contains the etc directory and the database you only need this tar archive backup to perform a complete restore.

Rotate Audit Log

Audit logs are written to the database. You can use pi-manage to perform a log rotation.

pi-manage rotate_audit

You can specify a highwatermark and a lowwatermark, age or a config file. Read more about it at Cleaning up entries.

API Keys

You can use pi-manage to create API keys. API keys can be used to

1. secure the access to the /validate/check API or
2. to access administrative tasks via the REST API.

You can create API keys for /validate/check using the command

pi-manage api createtoken -r validate

If you want to secure the access to /validate/check you also need to define a policy in scope authorizaion. See api_key_required.

If you wan to use the API key to automate administrative REST API calls, you can use the command:

pi-manage api createtoken -r admin

This command also generates an admin account name. But it does not create this admin account. You need to do so using pi-manage admin. You can now use this API key to enroll tokens as administrator.

Note

These API keys are not persistant. They are not stored in the privacyIDEA server. The API key is connected to the username, that is also generated. This means you have to create an administrative account with this very username to use this API key for this admin user. You also should set policies for this admin user, so that this API key has only restricted rights!

Note

The API key is valid for 365 days.

Policies

You can use pi-manage policy to enable, disable, create and delete policies. Using the sub commands p_export and p_import you can also export a backup of your policies and import this policy set later.

This could also be used to transfer the policies from one privacyIDEA instance to another.

Security Modules

Note

For a normal installation this section can be safely ignored.

privacyIDEA provides a security module that takes care of

• encrypting the token seeds,
• encrypting passwords from the configuration like the LDAP password,
• creating random numbers,
• and hashing values.

Note

The Security Module concept can also be used to add a Hardware Security Module to perform the above mentioned tasks.

Default Security Module

The default security module is implemented with the operating systems capabilities. The encryption key is located in a file enckey specified via PI_ENCFILE in the configuration file (The Config File).

This enckey contains three 32byte keys and is thus 96 bytes. This file has to be protected. So the access rights to this file are set accordingly.

In addition you can encrypt this encryption key with an additional password. In this case, you need to enter the password each time the privacyIDEA server is restarted and the password for decrypting the enckey is kept in memory.

The pi-manage Script contains the instruction how to encrypt the enckey

After starting the server, you can check, if the encryption key is accessible. To do so run:

privacyidea -U <yourserver> --admin=<youradmin> securitymodule

The output will contain "is_ready": True to signal that the encryption key is operational.

If it is not yet operational, you need to pass the password to the privacyIDEA server to decrypt the encryption key. To do so run:

privacyidea -U <yourserver> --admin=<youradmin> securitymodule  \
--module=default

Note

If the security module is not operational yet, you might get an error message “HSM not ready.”.

PKCS11 Security Module

The PKCS11 Security Module can be used to encrypt data with an hardware security module, that is connected via the PKCS11 interface. To encrypt and decrypt data you can use an RSA key pair that is stored on the HSM.

To activate this module add the following to the configuration file (The Config File)

PI_HSM_MODULE = “privacyidea.lib.security.pkcs11.PKCS11SecurityModule”

Additional attributes are

PI_HSM_MODULE_MODULE which takes the pkcs11 library. This is the full specified path to the shared object file in the file system.

PI_HSM_MODULE_KEY_ID is the key id (integer) on the HSM.

AES HSM Security Module

The AES Hardware Security Module can be used to encrypt data with an hardware security module (HSM) connected via the PKCS11 interface. This module allows to use AES keys stored in the HSM to encrypt and decrypt data.

This module uses three keys, similarly to the content of PI_ENCFILE, identified as token, config and value.

To activate this module add the following to the configuration file (The Config File)

PI_HSM_MODULE = “privacyidea.lib.security.aeshsm.AESHardwareSecurityModule”

Additional attributes are

PI_HSM_MODULE_MODULE which takes the pkcs11 library. This is the full specified path to the shared object file in the file system.

PI_HSM_MODULE_SLOT is the slot on the HSM where the keys are located (default: 1).

PI_HSM_MODULE_PASSWORD is the password to access the slot.

PI_HSM_MODULE_KEY_LABEL is the label prefix for the keys on the HSM (default: privacyidea). In order to locate the keys, the module will search for key with a label equal to the concatenation of this prefix, _ and the key identifier (respectively token, config and value).

PI_HSM_MODULE_KEY_LABEL_TOKEN is the label for token key (defaults to value based on PI_HSM_MODULE_KEY_LABEL setting).

PI_HSM_MODULE_KEY_LABEL_CONFIG is the label for config key (defaults to value based on PI_HSM_MODULE_KEY_LABEL setting).

PI_HSM_MODULE_KEY_LABEL_VALUE is the label for value key (defaults to value based on PI_HSM_MODULE_KEY_LABEL setting).

After installation you might want to take a look at First Steps.

Footnotes

[1]	https://launchpad.net/~privacyidea

[2]	https://github.com/privacyidea/privacyidea/tree/master/authmodules/simpleSAMLphp

First Steps

You installed privacyIDEA successfully according to Installation and created an administrator using the command pi-manage admin as e.g. described in Ubuntu Packages.

These first steps will guide you through the tasks of logging in to the management web UI, attaching your first users and enrolling the first token.

Login to the Web UI

privacyIDEA has only one login form that is used by administrators and normal users to login. Administrators will be able to configure the system and to manage all tokens, while normal users will only be able to manage their own tokens.

You should enter your username with the right realm. You need to append the realm to the username like username@realm.

Login for administrators

Administrators can authenticate at this login form to access the management UI.

Administrators are stored in the database table Admin and can be managed with the tool:

pi-manage admin ...

The administrator just logs in with his username.

Note

You can configure privacyIDEA to authenticate administrators against privacyIDEA itself, so that administrators need to login with a second factor. See SUPERUSER_REALM in inifile_superusers how to do this.

Login for normal users

Normal users authenticate at the login form to be able to manage their own tokens. By default users need to authenticate with the password from their user source.

E.g. if the users are located in an LDAP or Active Directory the user needs to authenticate with his LDAP/AD password.

But before a user can login, the administrator needs to configure realms, which is described in the next step Creating your first realm.

Note

The user my either login with his password from the userstore or with any of his tokens.

Note

The administrator may change this behaviour by creating an according policy, which then requires the user to authenticate against privacyIDEA itself. I.e. this way the user needs to authenticate with a second factor/token to access the self service portal. (see the policy section login_mode)

Creating your first realm

Note

When the administrator logs in and no useridresolver and no realm is defined, a popup appears, which asks you to create a default realm. During these first steps you may say “No”, to get a better understanding.

Users in privacyIDEA are read from existing sources. See Realms for more information.

In these first steps we will simply read the users from your /etc/passwd file.

Create a UserIdResolver

The UserIdResolver is the connector to the user source. For more information see UserIdResolvers.

• Go to Config -> Users to create a UserIdResolver.

Create the first UserIdResolver

• Choose New passwdresolver and
• Enter the name “myusers”.
• Save it.

Create the first UserIdResolver

You just created your first connection to a user source.

Create a Realm

User sources are grouped togeather to a so called “realm”. For more information see Realms.

• Go to Config -> Realms
• Enter “realm1” as the new realm name and select the priority 1.
• Check the resolver “myusers” to be included into this realm.
• Save it.

Create the first Realm

• Go to Users and you will see the users from the /etc/passwd.

The users from /etc/passwd

Congratulation! You created your first realm.

You are now ready to enroll a token to a user. Read Enrolling your first token.

Enrolling your first token

You may now enroll a new token. In this example we are using the Google Authenticator App, that you need to install on your smartphone.

• Go to Tokens -> Enroll Token

The Token Enrollment Dialog

• Select the username root. When you start typing “r”, “o”... the system will find the user root automatically.
• Enter a PIN. I entered “test” ...
• ... and click “Enroll Token”.

Enrollment Success

• After enrolling the token you will see a QR code, that you need to scan with the Google Authenticator App.
• Click on the serial number link at the top of the dialog.

Test the Token

• Now you see the token details.
• Left to the button “Test Token” you can enter the PIN and the OTP value generated by the Google Authenticator.
• Click the button “Test Token”. You should see a green “matching 1 tokens”.

Congratulations! You just enrolled your first token to a user.

Now you are ready to attach applications to privacyIDEA in order to add two factor authentication to those applications. To attach applications read the chapter Application Plugins.

You may also go on reading the chapter Configuration to get a deeper insight in the configuration possibilities.

After these first steps you will be able to start attaching applications to privacyIDEA in order to add two factor authentication to those applications. You can

• use a PAM module to authenticate with OTP at SSH or local login
• or the RADIUS plugin to configure your firewall or VPN to use OTP,
• or use an Apache2 plugin to do Basic Authentication with OTP.
• You can also setup different web applications to use OTP.

To attach applications read the chapter Application Plugins.

You may also go on reading the chapter Configuration to get a deeper insight in the configuration possibilities.

Configuration

The configuration menu can be used to define useridresolvers and realms, set the system config and the token config.

It also contains a shortcut to the policy tab (see Policies).

UserIdResolvers

Each organisation or company usually has its users managed at a central location. This is why privacyIDEA does not provide its own user management but rather connects to existing user stores.

UserIdResolvers are connectors to those user stores, the locations, where the users are managed. Nowadays this can be LDAP directories or especially Active Directory, some times FreeIPA or the Redhat 389 service. But classically users are also located in files like /etc/passwd on standalone unix systems. Web services often use SQL databases as user store.

Today with many more online cloud services SCIM is also an uprising protocol to access userstores.

privacyIDEA already comes with UserIdResolvers to talk to all these user stores:

• Flatfile resolver,
• LDAP resolver,
• SQL resolver,
• SCIM resolver.

Note

New resolver types (python modules) can be added easily. See the module section for this (UserIdResolvers).

You can create as many UserIdResolvers as you wish and edit existing resolvers. When you have added all configuration data, most UIs of the UserIdResolvers have a button “Test resolver”, so that you can test your configuration before saving it.

Starting with privacyIDEA 2.4 resolvers can be editable, i.e. you can edit the users in the user store. Read more about this at Manage Users.

Note

Using the policy authentication:otppin=userstore users can authenticate with the password from their user store, being the LDAP password, SQL password or password from flat file.

Flatfile resolver

Flatfile resolvers read files like /etc/passwd.

Note

The file /etc/passwd does not contain the unix password. Thus, if you create a flatfile resolver from this file the functionality with otppin=userstore is not available. You can create a flatfile with passwords using the tool privacyidea-create-pwidresolver-user.

Create a flat file like this:

privacyidea-create-pwidresolver-user -u user2 -i 1002 >> /your/flat/file

LDAP resolver

The LDAP resolver can be used to access any kind of LDAP service like OpenLDAP, Active Directory, FreeIPA, Penrose, Novell eDirectory.

LDAP resolver configuration

In case of Active Directory connections you might need to check the box No anonymous referral chasing. The underlying LDAP library is only able to do anonymous referral chasing. Active Directory will produce an error in this case [1].

The Server URI can contain a comma separated list of servers. The servers are used to create a server pool and are used with a round robin strategy [3].

Example:

ldap://server1, ldaps://server2:1636, server3, ldaps://server4

This will create LDAP requests to

• server1 on port 389
• server2 on port 1636 using SSL
• server3 on port 389
• server4 on port 636 using SSL.

The Bind Type with Active Directory can either be chosen as “Simple” or as “NTLM”.

Note

When using bind type “Simple” you need to specify the Bind DN like cn=administrator,cn=users,dc=domain,dc=name. When using bind type “NTLM” you need to specify Bind DN like DOMAINNAME\username.

The LoginName attribute is the attribute that holds the loginname. It can be changed to your needs.

Starting with version 2.20 you can provide a list of attributes in LoginName Attribute like:

sAMAccountName, userPrincipalName

This way a user can login with either his sAMAccountName or his principalName.

The searchfilter is used to list all possible users, that can be used in this resolver. The searchfilter is used for forward and backward search the object in LDAP.

The attribute mapping maps LDAP object attributes to user attributes in privacyIDEA. privacyIDEA knows the following attributes:

• phone,
• mobile,
• email,
• surname,
• givenname,
• password
• accountExpires.

The above attributes are used for privacyIDEA’s normal functionality and are listed in the userview. However, with a SAML authentication request user attributes can be returned. (see SAML Attributes). To return arbitrary attributes from the LDAP you can add additional keys to the attribute mapping with a key, you make up and the LDAP attribute like:

“homedir”: “homeDirectory”, “studentID”: “objectGUID”

“homeDirectory” and “objectGUID” being the attributes in the LDAP directory and “homedir” and “studentID” the keys returned in a SAML authentication request.

The MULTIVALUEATTRIBUTES config value can be used to specify a list of user attributes, that should return a list of values. Imagine you have a user mapping like { "phone" : "telephoneNumber", "email" : "mail", "surname" : "sn", "group": "memberOf"}. Then you could specify ["email", "group"] as the multi value attribute and the user object would return the emails and the group memberships of the user from the LDAP server as a list.

Note

If the MULTIVALUEATTRIBUTES is left blank the default setting is “mobile”. I.e. the mobile number will be returned as a list.

The MULTIVALUEATTRIBUTES can be well used with the samlcheck endpoint (see Validate endpoints) or with the policy add_user_in_response.

The UID Type is the unique identifier for the LDAP object. If it is left blank, the distinguished name will be used. In case of OpenLDAP this can be entryUUID and in case of Active Directory objectGUID. For FreeIPA you can use ipaUniqueID.

Note

The attributes entryUUID, objectGUID, and ipaUniqueID are case sensitive!

The option No retrieval of schema information can be used to disable the retrieval of schema information [4] in order to improve performance. This checkbox is deactivated by default and should only be activated after having ensured that schema information are unnecessary.

TLS certificates

Starting with privacyIDEA 2.18 in case of encrypted LDAPS connections privacyIDEA can verify the TLS certificate. (Python >= 2.7.9 required) To have privacyIDEA verify the TLS certificate you need to check the according checkbox.

You can specify a file with the trusted CA certificate, that signed the TLS certificate. The default CA filename is /etc/privacyidea/ldap-ca.crt and can contain a list of base64 encoded CA certificates. PrivacyIDEA will use the CA file if specifed. If you leave the field empty it will also try the system certificate store (/etc/ssl/certs/ca-certificates.crt or /etc/ssl/certs/ca-bundle.crt).

Modifying users

Starting with privacyIDEA 2.12 you can define the LDAP resolver as editable. I.e. you can create and modify users from within privacyIDEA.

There are two additional configuration parameters for this case.

DN Template defines how the DN of the new LDAP object should be created. You can use username, surname, givenname and basedn to create the distiguished name.

Examples:

CN=<givenname> <surname>,<basedn>

CN=<username>,OU=external users,<basedn>

uid=<username>,ou=users,o=example,c=com

Object Classes defines which object classes the user should be assigned to. This is a comma separated list. The usual object classes for Active Directory are

top, person, organizationalPerson, user, inetOrgPerson

Expired Users

You may set

“accountExpires”: “accountExpires”

in the attribute mapping for Microsoft Active Directories. You can then call the user listing API with the parameter accountExpires=1 and you will only see expired accounts.

This functionality is used with the script privacyidea-expired-users.

SQL resolver

The SQL resolver can be used to retrieve users from any kind of SQL database like MySQL, PostgreSQL, Oracle, DB2 or sqlite.

SQL resolver configuration

In the upper frame you need to configure the SQL connection. The SQL resolver uses SQLAlchemy internally. In the field Driver you need to set a driver name as defined by the SQLAlchemy dialects like “mysql” or “postgres”.

In the SQL attributes frame you can specify how the users are identified.

The Database table contains the users.

Note

At the moment only one table is supported, i.e. if some of the user data like email address or telephone number is located in a second table, those data can not be retrieved.

The Limit is the SQL limit for a userlist request. This can be important if you have several thousand user entries in the table.

The Attribute mapping defines which table column should be mapped to which privayIDEA attribute. The known attributes are:

• userid (mandatory),
• username (mandatory),
• phone,
• mobile,
• email,
• givenname,
• surname,
• password.

The password attribute is the database column that contains the user password. This is used, if you are doing user authentication against the SQL database.

Note

There is no standard way to store passwords in an SQL database. There are several different ways to do this. privacyIDEA supports the most common ways like Wordpress hashes starting with $P or $S. Secure hashes starting with {SHA} or salted secure hashes starting with {SSHA}, {SSHA256} or {SSHA512}. Password hashes of length 64 are interpreted as OTRS sha256 hashes.

You can mark the users as Editable. The Password_Hash_Type can be used to determine wich hash algorithm should be used, if a password of an editable user is written to the database.

You can add an additional Where statement if you do not want to use all users from the table.

The poolSize and poolTimeout determine the pooling behaviour. The poolSize (default 5) determine how many connections are kept open in the pool. The poolTimeout (default 10) specifies how long the application waits to get a connection from the pool.

Note

The Additional connection parameters refer to the SQLAlchemy connection but are not used at the moment.

SCIM resolver

SCIM is a “System for Cross-domain Identity Management”. SCIM is a REST-based protocol that can be used to ease identity management in the cloud.

The SCIM resolver is tested in basic functions with OSIAM [2], the “Open Source Idenitty & Access Management”.

To connect to a SCIM service you need to provide a URL to an authentication server and a URL to the resource server. The authentication server is used to authenticate the privacyIDEA server. The authentication is based on a client name and the Secret for this client.

Userinformation is then retrieved from the resource server.

The available attributes for the Attribute mapping are:

• username (mandatory),
• givenname,
• surname,
• phone,
• mobile,
• email.

User Cache

privacyIDEA does not implement local user management by design and relies on UserIdResolvers to connect to external user stores instead. Consequently, privacyIDEA queries user stores quite frequently, e.g. to resolve a login name to a user ID while processing an authentication request, which may introduce a significant slowdown. In order to optimize the response time of authentication requests, privacyIDEA 2.19 introduces the user cache which is located in the local database. It can be enabled in the system configuration (see User Cache).

A user cache entry stores the association of a login name in a specific UserIdResolver with a specific user ID for a predefined time called the expiration timeout, e.g. for one week. The processing of further authentication requests by the same user during this timespan does not require any queries to the user store, but only to the user cache.

The user cache should only be enabled if the association of users and user ID is not expected to change often: In case a user is deleted from the user store, but can still be found in the user cache and still has assigned tokens, the user will still be able to authenticate during the expiration timeout! Likewise, any changes to the user ID will not be noticed by privacyIDEA until the corresponding cache entry expires.

Expired cache entries are not deleted from the user cache table automatically. Instead, the tool privacyidea-usercache-cleanup should be used to delete expired cache entries from the database, e.g. in a cronjob.

However, cache entries are removed at some defined events:

• If a UserIdResolver is modified or deleted, all cache entries belonging to this resolver are deleted.
• If a user is modified or deleted in an editable UserIdResolver, all cache entries belonging to this user are deleted.

Note

Realms with multiple UserIdResolvers are a special case: If a user userX tries to authenticate in a realm with two UserIdResolvers resolverA (with highest priority) and resolverB, the user cache is queried to find the user ID of userX in the UserIdResolver resolverA. If the cache contains no matching entry, resolverA itself is queried for a matching user ID! Only if resolverA does not find a corresponding user, the user cache is queried to determine the user ID of userX in resolverB. If no matching entry can be found, resolverB is queried.

Footnotes

[1]	http://blogs.technet.com/b/ad/archive/2009/07/06/referral-chasing.aspx

[2]	http://www.osiam.org

[3]	https://github.com/cannatag/ldap3/blob/master/docs/manual/source/servers.rst#server-pool

[4]	http://ldap3.readthedocs.io/schema.html

Realms

Users need to be in realms to have tokens assigned. A user, who is not member of a realm can not have a token assigned and can not authenticate.

You can combine several different UserIdResolvers (see UserIdResolvers) into a realm. The system knows one default realm. Users within this default realm can authenticate with their username.

Users in realms, that are not the default realm, need to be additionally identified. Therefor the users need to authenticate with their username and the realm like this:

user@realm

List of Realms

The realms dialog gives you a list of the already defined realms.

It shows the name of the realms, whether it is the default realm and the names of the resolvers, that are combined to this realm.

You can delete or edit an existing realm or create a new realm.

Edit Realm

Each realm has to have a unique name. The name of the realm is case insensitive. If you create a new realm with the same name like an existing realm, the existing realm gets overwritten.

If you click Edit Realm you can select which userresolver should be contained in this realm. A realm can contain several resolvers.

Edit a realm

Resolver Priority

Within a realm you can give each resolver a priority. The priority is used to find a user that is located in several resolvers. If a user is located in more than one resolver, the user will be taken from the resolver with the lowest number in the priority.

Priorities are numbers between 1 and 999. The lower the number the higher the priority.

Example:

A user “administrator” is located in a resolver “users” which contains all Active Directory users. And the “administrator” is located in a resolver “admins”, which contains all users in the Security Group “Domain Admins” from the very same domain. Both resolvers are in the realm “AD”, “admins” with priority 1 and “users” with priority 2.

Thus the user “administrator@AD” will always resolve to the user located in resolver “admins”.

This is useful to create policies for the security group “Domain Admins”.

Note

A resolver has a priority per realm. I.e. a resolver can have a different priority in each realm.

Autocreate Realm

If you have a fresh installation, no resolver and no realm is defined. To get you up and running faster, the system will ask you, if it should create the first realm for you.

If you answer “yes”, it will create a resolver named “deflocal” that contains all users from /etc/passwd and a realm named “defrealm” with this very resolver.

Thus you can immediately start assigning and enrolling tokens.

If you check “Do not ask again” this will be stored in a cookie in your browser.

Note

The realm “defrealm” will be the default realm. So if you create a new realm manually and want this new realm to be the default realm, you need to set this new realm to be default manually.

System Config

The system configuration has three logical topics: Settings, token default settings and GUI settings.

The system config

Settings

Split @ Sign

splitAtSign defines if the username like user@company given during authentication should be split into the loginname user and the realm name company. In most cases this is the wanted behaviour.

But given your users log in with email addresses like user@gmail.com and otheruser@outlook.com you probably do not want to split.

SAML Attributes

Return SAML attributes defines if during an SAML authentication request additional SAML attributes should be returned. Usually an authentication response only returns true or false.

The SAML attributes are the known attributes that are defined in the attribute mapping e.g. of the LDAP resolver like email, phone, givenname, surname or any other attributes you fetch from the LDAP directory. For more information read LDAP resolver.

In addition you can set the parameter ReturnSamlAttributesOnFail. In this case the response contains the SAML attributes of the user, even if the user failed to authenticate.

FailCounterIncOnFalsePin

If during authentication the given PIN matches a token but the OTP value is wrong the failcounter of the tokens for which the PIN matches, is increased. If the given PIN does not match any token, by default no failcounter is increased. The later behaviour can be adapted by FailCounterIncOnFalsePin. If FailCounterIncOnFalsePin is set and the given OTP PIN does not match any token, the failcounter of all tokens is increased.

Automatically clearing Failcounter

If the failcounter reaches the maximum the token gets a timestamp, when the max fail count was reached. A successful authentication after the specified a amount of minutes in failcounter_clear_timeout will clear the failcounter again and the user can authenticate.

A “0” means automatically clearing the fail counter is not used.

Also see How to mitigate brute force and lock tokens.

Prepend PIN

PrependPin defines if the OTP PIN should be given in front (“pin123456”) or in the back (“12345pin”) of the OTP value.

AutoResync

Auto resync defines if the system should try to resync a token if a user provides a wrong OTP value. AutoResync works like this:

• If the counter of a wrong OTP value is within the resync window, the system remembers the counter of the OTP value for this token in the token info field otp1c.
• Now the user needs to authenticate a second time within auto resync timeout with the next successive OTP value.
• The system checks if the counter of the second OTP value is the successive value to otp1c.
• If it is, the token counter is set and the user is successfully authenticated.

Note

AutoResync works for all HOTP and TOTP based tokens including SMS and Email tokens.

User Cache

The setting User Cache expiration in seconds is used to enable the user cache and configure its expiration timeout. If its value is set to 0 (which is the default value), the user cache is disabled. Otherwise, the value determines the time in seconds after which entries of the user cache expire. For more information read User Cache.

Note

If the user cache is already enabled and you increase the expiration timeout, expired entries that still exist in the user cache could be considered active again!

Override Authorization Client

Override Authorization client is important with client specific policies (see Policies) and RADIUS servers or other proxies. In case of RADIUS the authenticating client for the privacyIDEA system will always be the RADIUS server, which issues the authentication request. But you can allow the RADIUS server IP to send another client information (in this case the RADIUS client) so that the policy is evaluated for the RADIUS client. Such a proxy or RADIUS server may add the API parameter client with a new IP address.

This field takes a comma separated list of IP Networks mapping to other IP Networks.

Examples

10.1.2.0/24 > 192.168.0.0/16*

Proxies in the sub net 10.1.2.0/24 may mask as client IPs 192.168.0.0/16. In this case the policies for the corresponding client in 192.168.x.x apply.

172.16.0.1

The proxy 172.16.0.1 may mask as any arbitrary client IP.

10.0.0.18 > 10.0.0.0/8

The proxy 10.0.0.18 may mask as any client in the subnet 10.x.x.x.

Token default settings

Reset Fail Counter

DefaultResetFailCount will reset the failcounter of a token if this token was used for a successful authentication. If not checked, the failcounter will not be resetted and must be resetted manually.

Note

The following settings are token specific value which are set during enrollment. If you want to change this value of a token later on, you need to change this at the tokeninfo dialog.

Maximum Fail Counter

DefaultMaxFailCount is the maximum failcounter a token way get. If the failcounter exceeds this number the token can not be used unless the failcounter is resetted.

Note

In fact the failcounter will only increase till this maxfailcount. Even if more failed authentication request occur, the failcounter will not increase anymore.

Sync Window

DefaultSyncWindow is the window how many OTP values will be calculated during resync of the token.

OTP Length

DefaultOtpLen is the length of the OTP value. If no OTP length is specified during enrollment, this value will be used.

Count Window

DefaultCountWindow defines how many OTP values will be calculated during an authentication request.

Challenge Validity Time

DefaultChallengeValidityTime is the timeout for a challenge response authentication. If the response is set after the ChallengeValidityTime, the response is not accepted anymore.

SerialLength

The default length of generated serial numbers is an 8 digit hex string. If you need another length, it can be configured in the database table Config with the key word SerialLength.

Tokens

Supported Tokens

privacyIDEA supports a wide variety of tokens by different hardware vendors. It also supports token apps on the smartphone.

Tokens not listed, will be probably supported, too, since most tokens use standard algorithms.

If in doubt drop your question on the mailing list.

Hardware Tokens

The following hardware tokens are known to work well.

Yubikey. The Yubikey is supported in all modes: AES (Yubikey), HOTP and Yubico Cloud. You can initialize the Yubikey yourself, so that the secret key is not known to the vendor.

eToken Pass. The eToken Pass is a push button token by SafeNet. It can be initialized with a special hardware device. Or you get a seed file, that you need to import to privacyIDEA. The eToken Pass can run as HOTP or TOTP token.

eToken NG OTP. The eToken NG OTP is a push button token by SafeNet. As it has a USB connector, you can initialize the token via the USB connector. Thus the hardware vendor does not know the secret key.

DaPlug. The DaPlug token is similar to the Yubikey and can be initialized via the USB connector. The secret key is not known to the hardware vendor.

Smartdisplayer OTP Card. This is a push button card. It features an eInk display, that can be read very good in all light condition at all angles. The Smartdisplayer OTP card is initialized at the factory and you get a seed file, that you need to import to privacyIDEA.

Feitian. The C100 and C200 tokens are classical, reasonably priced push button tokens. The C100 is an HOTP token and the C200 a TOTP token. These tokens are initialized at the factory and you get a seed file, that you need to import to privacyIDEA.

U2F. The Yubikey and the Daplug token are known U2F devices to work well with privacyIDEA. See U2F.

Smartphone Apps

privacyIDEA Authenticator. Our own privacyIDEA Authenticator is based on the concept of the Google Authenticator and works with the usual QR Code key URI enrollment. But on top it also allows for a more secure enrollment process (See Two Step Enrollment). It can be used for HOTP and TOTP.

Google Authenticator. The Google Authenticator is working well in HOTP and TOTP mode. If you choose “Generate OTP Key on the Server” during enrollment, you can scan a QR Code with the Google Authenticator. See Enrolling your first token to learn how to do this.

FreeOTP. privacyIDEA is known to work well with the FreeOTP App. The FreeOTP App is a TOTP token. So if you scan the QR Code of an HOTP token, the OTP will not validate.

mOTP. Several mOTP Apps like “Potato”, “Token2” or “DroidOTP” are supported.

Supported Tokentypes

At the moment the following tokentypes are supported:

• HOTP - event based One Time Password tokens based on RFC4225.
• TOTP - time based One Time Password tokens based on RFC6238.
• mOTP - time based One Time Password tokens for mobile phones based on an a public Algorithm.
• Paper Token - event based One Time Password tokens that get you list of one time passwords on a sheet of paper.
• Questionnaire Token - A token that contains a list of answered questions. During authentication a random question is presented as challenge from the list of answered questions is presented. The user must give the right answer.
• EMail - A token that sends the OTP value to the EMail address of the user.
• Four Eyes - Meta token that can be used to create a Two Man Rule.
• password - A password token used for losttoken scenario.
• Registration - A special token type used for enrollment scenarios (see Registration Code).
• Simple Pass - A token that only consists of the Token PIN.
• Certificates - A token that represents a client certificate.
• SSH Keys - An SSH public key that can be managed and used in conjunction with the Client machines concept.
• Remote - A virtual token that forwards the authentication request to another privacyIDEA server.
• RADIUS - A virtual token that forwards the authentication request to a RADIUS server.
• SMS - A token that sends the OTP value to the mobile phone of the user.
• Spass - Simple Pass Token - The simple pass token. A token that has no OTP component and just consists of the OTP pin or (if otppin=userstore is set) of the userstore password.
• TiQR - A Smartphone token that can be used to login by only scanning a QR code.
• OCRA - A basic OATH Challenge Response token.
• U2F - A U2F device as specified by the FIDO Alliance. This is a USB device to be used for challenge response authentication.
• VASCO - The proprietary VASCO token.
• Yubico - A Yubikey hardware that authenticates against the Yubico Cloud service.
• Yubikey - A Yubikey hardware initialized in the AES mode, that authenticates against privacyIDEA.
• Daplug - A hardware OTP token similar to the Yubikey.

The Tokentypes:

Four Eyes

Starting with version 2.6 privacyIDEA supports 4 Eyes Token. This is a meta token, that can be used to define, that two or more token must be used to authenticate. This way, you can set up a “two man rule”.

You can define, from which realm how many unique tokens need to be present, when authenticating:

Enroll a 4 eyes token

In this example authentication will only be possbile if at least two tokens from realm2 and one token from realm sqlite are present.

Authentication is done by concatenating the OTP PINs and the OTP values of all tokens. The concatenation is split by the separator character.

It does not matter, in which order the tokens from the realms are entered.

Example

Authentication as:

username: "root@r2"
password: "pin123456 secret789434 key098123"

The three blocks separated by the blank are checked, if they match tokens in the realms realm2 and sqlite.

The response looks like this in case of success:

{
  "detail": {
    "message": "matching 1 tokens",
    "serial": "PI4E000219E1",
    "type": "4eyes"
  },
  "id": 1,
  "jsonrpc": "2.0",
  "result": {
    "status": true,
    "value": true
  },
  "version": "privacyIDEA 2.6dev0",
  "versionnumber": "2.6dev0"
}

In case of a failed authentication the response looks like this:

{
  "detail": {
    "foureyes": "Only found 0 tokens in realm themis",
    "message": "wrong otp value",
    "serial": "PI4E000219E1",
    "type": "4eyes"
  },
  "id": 1,
  "jsonrpc": "2.0",
  "result": {
    "status": true,
    "value": false
  },
  "version": "privacyIDEA 2.6dev0",
  "versionnumber": "2.6dev0"
}

Note

The 4Eyes Token verifies that unique tokens from each realm are used. I.e. if you require 2 tokens from a realm, you can not use the same token twice.

Warning

But it does not verify, if these two unique tokens belong to the same user. Thus you should create a poliy, that in such a realm a user may only have on token.

Certificates

Starting with version 2.3 privacyIDEA supports certificates. A user can

• upload a certificate request,
• upload a certificate or
• he can generate a certificate request in the browser.

privacyIDEA does not sign certificate requests itself but connects to existing certificate authorities. To do so, you need to define CA Connectors.

Certificates are attached to the user just like normal tokens. One token of type certificate always contains only one certificate.

If you have defined a CA connector you can upload a certificate signing request (CSR) via the Token Enroll Dialog in the WebUI.

Upload a certificate signing request

You need to choose the CA connector. The certificate will be signed by the CA accordingly. Just like all other tokens the certificate token can be attached to a user.

Generating Signing Requests

You can also generate the signing request directly in your browser.

Note

This uses the keygen HTML-tag that is not supported by the Internet Explorer!

Generate a certificate signing request

When generating the certificate signing request this way the RSA keypair is generated on the client side in the browser.

The certificate is signed by the CA connected by the chosen CA connector.

Download or install the client certificate

Afterwards the user can install the certificate into the browser.

Note

By requiring OTP authentication for the users to login to the WebUI (see login_mode) you can have two factor authentication required for the user to be allowed to enroll a certificate.

EMail

The token type email sends the OTP value in an EMail to the user. You can configure the EMail server in Email OTP Token.

Enroll an EMail token

When enrolling an EMail token, you only need to specify the email address of the user.

The EMail token is a challenge response token. I.e. when using the OTP PIN in the first authentication request, the sending of the EMail will be triggered and in a second authentication request the OTP value from the EMail needs to be presented.

For a more detailed insight see the code documentation Email Token.

HOTP

The HOTP token is - together with the TOTP - the most common token. The HOTP Algorithm is defined in RFC4225. The HOTP token is an event base token. The HOTP algorithm has some parameter, like if the generated OTP value will be 6 digits or 8 digits or if the SHA1 oder the SHA256 hashing algorithm is used.

Hardware tokens

There are many token vendors out there who are using the official algorithm to build and sell hardware tokens. You can get HOTP based hardware tokens in different form factors, as a normal key fob for your key ring or as a display card for your purse.

Preseeded or Seedable

Usually the hardware tokens like keyfobs or display cards contain a secret key that was generated and implanted at the vendors factory. The vender ships the tokens and a seed file.

Warning

In this case privacyIDEA can not guarantee that the secret seed of the token is unique and if you are using a real strong factor.

privacyIDEA also supports the following seedable HOTP tokens:

• SafeNet eToken NG OTP
• SafeNet eToken Pass
• Yubikey in OATH mode
• Daplug

Those tokens can be initialized by privacyIDEA. Thus you can be sure, that only you are in possession of the secret seed.

Experiences

The above mentioned hardware tokens are known to play well with privacyIDEA. In theory all OATH/HOTP tokens should work well with privacyIDEA. However, there are good experiences with Smartdisplayer OTP cards [1] and Feitian C200 [2] tokens.

Software tokens

Besides the hardware tokens there are also software tokens, implemented as Apps for your smartphone. These software tokens allow are seedable, so there is no vendor, knowing the secret seed of your OTP tokens.

But software tokens are software after all on device prone to security issues.

Experiences

The Google Authenticator can be enrolled easily in HOTP mode using the QR-Code enrollment Feature.

The Google Authenticator is available for iOS, Android and Blackberry devices.

Enrollment

Default settings for HOTP tokens can be configured at HOTP Token Config.

Enroll an HOTP token

During enrollment you can choose, if the server should generate the key or if you have a key, that you can enter into the enrollment page.

As mentioned earlier, you can also choose the OTP length and the hash algoriothm.

If the server generated the secret seed, you can scan the QR-Code

After enrolling the token, the QR-Code, containing the secret seed, is displayed, so that you can scan this with your smartphone and import it to your app.

[1]	https://netknights.it/en/produkte/smartdisplayer/

[2]	https://netknights.it/en/produkte/oath-hotptotp/

OCRA

Starting with version 2.20 privacyIDEA supports common OCRA tokens. OCRA tokens can not be enrolled via the UI but need to be imported via a seed file. The OATH CSV seed file would look like this:

<serial>, <seed>, ocra, <ocrasuite>

The OCRA token is a challenge/response token. So the first authentication request issues a challenge. This challenge is the input for the response of the OCRA token.

For more information see OCRA Token.

DisplayTAN token

privacyIDEA supports the DisplayTAN [1], which can be used for securing banking transactions. The OCRA Algorithm is used to digitally sign transaction data. The transcation data can be verified by the user on an external banking card. All cryptographical processes are running on the external card, so that an attacker can not interfere with the user’s component.

The DisplayTAN cards would be imported into privacyIDEA using the token import.

A banking website will use the Validate endpoints API.

The first call will trigger the challenge response mechanism. The first call needs to contain the transaction data: the recipient’s account number and amount of money to transfer:

<account>~<amount>~

Please note the tilde:

POST https://privacyidea.example.com/validate/check

pass=pin
serial=ocra1234
challenge=1234567890~423,40~
addrandomchallenge=20
hashchallenge=sha1

This will result in a response like this:

{
  "jsonrpc": "2.0",
  "signature": "128057011582042...408",
  "detail": {
             "multi_challenge": [
                {
                 "attributes": {
                 "qrcode": "data:image/png;base64, iVBORw0KG..RK5CYII=",
                 "original_challenge": "83507112  ~320,
               00~cfbGSopfdDROOMjeu3IR",
                 "challenge": "f8a1818f35ae0cc64fe8a191961ec829487dfa82"
                 },
                 "serial": "ocra1234",
                 "transaction_id": "05221757445370623976"
                }
             ],
             "threadid": 139847557760768,
             "attributes": {
             "qrcode": "data:image/png;base64, iVBO...CYII=",
             "original_challenge": "83507112  ~320,00~cfbGSopfdDROOMjeu3IR",
             "challenge": "f8a1818f35ae0cc64fe8a191961ec829487dfa82"
             },
             "message": "Please answer the challenge",
             "serial": "ocra1234",
             "transaction_id": "05221757445370623976"
  },
  "versionnumber": "2.20.dev2",
  "version": "privacyIDEA 2.20.dev2",
  "result": {
             "status": true,
             "value": false
  },
  "time": 1504005837.417481,
  "id": 1
}

Note

The response also contains the QR code. The banking website should show the QR code, so that the user can scan it with the DisplayTAN App to transfer the data to the card.

The user can verify the data on the card and transaction data will be digitally signed on the card. The card will calculate an OTP value for this very transaction.

The banking website can now send the OTP value to privacyIDEA to check, if the user authorized the correct transaction data. The banking site will issue this request:

POST https://privacyidea.example.com/validate/check

serial=ocra1234
transaction_id=05221757445370623976
pass=54006635

privcyIDEA will respond with a usual authentication response:

{
 "jsonrpc": "2.0",
 "signature": "162....2454851",
 "detail": {
            "message": "Found matching challenge",
            "serial": "ocra1234",
            "threadid": 139847549368064
           },
 "versionnumber": "2.20.dev2",
 "version": "privacyIDEA 2.20.dev2",
 "result": {
            "status": true,
            "value": true
 },
 "time": 1504005901.823667,
 "id": 1
}

[1]	http://www.display-tan.com/

Paper Token

The token type paper lets you print out a list of OTP values, which you can use to authenticate and cross of the list.

The paper token is based on the HOTP. I.e. you need to use one value after the other.

Customization

CSS

You can customize the look and feel of the printed paper token. You may change the style sheep papertoken.css which is only loaded for printing.

Header and Footer

Then you may add a header in front and a footer behind the table containing the OTP values.

Create the files

• static/customize/views/includes/token.enrolled.paper.top.html
• static/customize/views/includes/token.enrolled.paper.bottom.html

to display the contents before (top) and behind (bottom) the table.

Within these html templates you may use angular replacements. To get the serial number of the token use

{{ tokenEnrolled.serial }}

to get the name and realm of the user use

{{ newUser.user }} {{ newUser.realm }}

A good example for the token.enrolled.paper.top.html is:

<h1>{{ enrolledToken.serial }}</h1> <p>

Please use the OTP values of your paper token in order one after the other. You may scratch of or otherwise mark used values.

</p>

A good example for the token.enrolled.paper.bottom.html is:

<p>

The paper token is a weak second factor. Please assure, that noone gets hold of this paper and can make a copy of it.

</p> <p>

Store it at a safe location.

</p>

Note

You can change the directory static/customize to a URL that fits your needs the best by defining a variable PI_CUSTOMIZATION in the file pi.cfg. This way you can put all modifications in one place apart from the original code.

OTP Table

If you want to change the complete layout of the table you need to overwrite the file static/components/token/views/token.enrolled.paper.html. The scope variable {{ enrolledToken.otps }} contains an object with the complete OTP value list.

Questionnaire Token

The administrator can define a list of questions and also how many answers to the questions a user needs to define.

During enrollment of such a question type token the user answers at least as many questions as specified with answers only he knows.

This token is a challenge response token. During authentication the user must give the token PIN and the a random question from the answered question is chosen. The user has to answer with the same answer he defined earlier.

Note

If the administrator changes the questions _after_ a token was enrolled, the enrolled token still works with the old questions and answers. I.e. an enrolled token is not affected by changing the questions by the administrator.

RADIUS

The token type RADIUS forwards the authentication request to a RADIUS Server.

When forwarding the authentication request, you can change the username and mangle the password.

Enroll a RADIUS token

Check the PIN locally

If checked, the PIN of the token will be checked on the local server. If the PIN matches only the remaining part of the issued password will be sent to the RADIUS server.

RADIUS Server

The RADIUS server, to which the authentication request will be forwarded. You can specify the port like my.radius.server:1812.

RADIUS User

When forwarding the request to the RADIUS server, the authentication request will be issued for this user. If the user is left empty, the RADIUS request will be sent with the same user.

RADIUS Secret

The RADIUS secret for this RADIUS client.

Note

Using the RADIUS token you can design migration scenarios. When migrating from other (proprietary) OTP solutions, you can enroll a RADIUS token for the users. The RADIUS token points to the RADIUS server of the old solution. Thus the user can authenticate against privacyIDEA with the old, proprietary token, till he is enrolled a new token in privacyIDEA. The interesting thing is, that you also get the authentication request with the proprietary token in the audit log of privacyIDEA. This way you can have a scenario, where users are still using old tokens and other users are already using new (privacyIDEA) tokens. You will see all authentication requests in the pricacyIDEA system.

Registration

(See Registration Code)

The registration token can be used to create a registration code for a user. This registration code can be sent via postal mail to the user, so that the user can use this registration code as a second factor to login to a portal.

After a one single use, the registration code is deleted and can not be used a second time.

Note

The registration code can only be enrolled via the API to provide automated smooth workflow to your needs.

For a more detailed insight see the code documentation Registration Code Token.

Remote

The token type remote forwards the authentication request to another privacyIDEA Server.

When forwarding the authentication request, you can

• change the username
• change the resolver
• change the realm
• change the serial number

and mangle the password.

Enroll a Remote token

Check the PIN locally

If checked, the PIN of the token will be checked on the local server. If the PIN matches only the remaining part of the issued password will be sent to the remote privacyIDEA server.

Remote Server

The privacyIDEA server, to which the authentication request will be forwarded. The path /validate/check will be added automatically. So a sensible input would be https://my.other.server/.

Remote Serial

If the Remote Serial is specified the given password will be checked against the serial number on the remote privacyIDEA server. Usernames will be ignored.

Remote User

When forwarding the request to the remote server, the authentication request will be issued for this user.

Remote Realm

When forwarding the request to the remote server, the authentication request will be issued for this realm.

Remote Resolver

When forwarding the request to the remote server, the authentication request will be issued for this resolver.

Note

You can use Remote Serial to forward the request to a central privacyIDEA server, that only knows tokens but has no knowledge of users. Or you can use Remote Serial to forward the request to an existing to on localhost thus adding a second user to the same token.

SMS

The token type sms sends the OTP value via an SMS service. You can configure the SMS service in SMS OTP Token.

Enroll an SMS token

When enrolling an SMS token, you only need to specify the mobile phone number.

SMS token is a challenge response token. I.e. when sending the OTP PIN in the first authentication request, the sending of the SMS will be triggered and in a second authentication request the OTP value from the SMS needs to be presented.

For a more detailed insight see the code documentation SMS Token.

Spass - Simple Pass Token

The OTP component of the spass token is always true. Thus the user only needs to provide the OTP pin or the userstore password - depending on the policy settings.

For a more detailed insight see the code documentation SPass Token.

SSH Keys

The token type sshkey is the public SSH key, that you can upload and assign to a user. The SSH key is only used for the application type SSH in conjunction with the Client machines concept.

A user or the administrator can upload the public SSH key and assign to a user.

Enroll an SSH key token

Paste the SSH key into the text area. The comment in the SSH key will be used as token comment. You can assign the SSH key to a user and then use the SSH key in Application Definitions SSH.

Note

This way you can manage SSH keys centrally, as you do not need to distribute the SSH keys to all machines. You rather store the SSH keys centrally in privacyIDEA and use privacyidea-authorizedkeys to fetch the keys in real time during the login process.

TiQR

Starting with version 2.6 privacyIDEA supports the TiQR token. The TiQR token is a smartphone token, that can be used to login by only scanning a QR code.

The token is also enrolled by scanning a QR code.

Choose a user for the TiQR token

You can only enroll a TiQR token, when a user is selected.

Note

You can not enroll a TiQR token without assign the token to a user.

For more technical information about the TiQR token please see TiQR Token.

TOTP

The TOTP token is - together with the HOTP - the most common token. The TOTP Algorithm is defined in RFC6238. The TOTP token is a time based token. Roughly speaking the TOTP algorithm is the same algorithm like the HOTP, where the event based counter is replaced by the unix timestamp.

The TOTP algorithm has some parameter, like if the generated OTP value will be 6 digits or 8 digits or if the SHA1 oder the SHA256 hashing algorithm is used and the timestep being 30 or 60 seconds.

Hardware tokens

The information about preseeded token and seedable tokens is the same as described in the section about HOTP.

The only available seedable pushbutton TOTP token is the SafeNet eToken Pass. The Yubikey can be used as a TOTP token, but only in conjunction with a smartphone app, since the yubikey has not its own clock.

Software tokens

Experiences

The Google Authenticator and the FreeOTP token can be enrolled easily in TOTP mode using the QR-Code enrollment Feature.

The Google Authenticator is available for iOS, Android and Blackberry devices.

Enrollment

Default settings for TOTP tokens can be configured at TOTP Token Config.

The enrollment is the same as described in HOTP. However, when enrolling TOTP token, you can specify some additional parameters.

Enroll an TOTP token

U2F

Starting with version 2.7 privacyIDEA supports U2F tokens. The administrator or the user himself can register a U2F device and use this U2F token to login to the privacyIDEA web UI or to authenticate at applications.

When enrolling the token a key pair is generated and the public key is sent to privacyIDEA. During this process the user needs to prove that he is present by either pressing the button (Yubikey) or by replugging the device (Plug-up token).

The device is identified and assigned to the user.

Note

This is a normal token object which can also be reassigned to another user.

Note

As the key pair is only generated virtually, you can register one physical device for several users.

For configuring privacyIDEA for the use of U2F token, please see U2F Token Config.

For further details and for information how to add this to your application you can see the code documentation at U2F Token.

VASCO

Starting with version 2.22 privacyIDEA supports VASCO tokens.

VASCO OTP tokens are a proprietary OTP token. You can import the VASCO blobs from a CSV file or you the administrator can enroll a single VASCO token.

Note

privacyIDEA uses a proprietary VASCO library vacman to verify the OTP values. Please note that you need to license this library from VASCO Data Security N.V. directly. The privacyIDEA project does not provide this library.

Yubico

The token type yubico authenticates against the Yubico Cloud mode. You need to configure this at Yubico Cloud mode.

Enroll a Yubico token

The token is enrolled by simply saving the Yubikey token ID in the token object. You can either enter the 12 digit ID or you can simply press the Yubikey button in the input field, which will also assign the token.

Yubikey

The Yubikey is initialized with privacyIDEA and works in Yubicos own AES mode. It outputs a 44 character OTP value, consisting of a 12 character prefix and a 32 character OTP. But in contrast to the Yubico Cloud mode, in this mode the secret key is contained within the token and your own privacyIDEA installation.

If you have the time and care about privacy, you should prefer the Yubikey AES mode over the Yubico Cloud mode.

There are three possible ways to enroll a Yubikey token.

Note

We recommend that you use the privacyidea command line client, to initialize the Yubikeys. You can use the mass enrollment, which eases the process of initializing a whole bunch of tokens.

Run the command like this:

privacyidea -U https://your.privacyidea.server -a admin token \
yubikey_mass_enroll --yubimode YUBICO

This command initializes the token and stores the AES secret and prefix in privacyidea, so the token is immediatly useful. You can choose the slot with --yubislot. For further help call privcyidea yubikey_mass_enroll with the --help option.

The second way to enroll a yubikey token is also using yubikey_mass_enroll, but with the option --filename to write to token configuration into the specified file. The resulting file can then be imported into privacyidea: Select Tokens -> Import Tokens, select “OATH CSV” and the file you just created.

Using the yubikey personalization GUI

Third and last you can use the privacyIDEA Web UI to enroll a Yubikey AES mode token, if you have initialized the yubikey with the external ykpersonalize tool.

Use the yubikey-personalization-gui to initialize the yubikey

When using the yubikey personalization GUI you need to copy the value of “Secret Key (16 bytes Hex)”. This is the secret OTP key, which you need to copy and paste in the field “OTP Key” in the privacyIDEA Web UI. (Remove possible white spaces!)

Enroll a Yubikey AES mode token

In the field “Test Yubikey” push the Yubikey button. This will grab the yubikey’s public identifier and also determine the lenght of the otp value. The field OTP value is automatically filled.

Redirect api url to privacyideas /ttype/yubikey

Yubico servers use /wsapi/2.0/verify as the path in the validation URL. Some tools (e.g. Kolab 2fa) let the user/admin change the api host, but not the rest of the URL. Let’s redirect the api URL to privacyideas /ttype/yubikey - you’ll need to enable the following two lines in /etc/apache2/site-enabled/privacyidea.conf:

RewriteEngine on RewriteRule “^/wsapi/2.0/verify” “/ttype/yubikey” [PT]

If you use nginx there is a similar line provided as a comment to the nginx configuration as well.

Token configuration

Each token type can provide its own configuration dialog.

In this configuration dialog you can define default values for these token types.

Token Configuration: SMS

Email OTP Token

Email Token configuration

The Email OTP token creates a OTP value and sends this OTP value to the email address of the uses. The Email can be triggered by authenticating with only the OTP PIN:

First step

In the first step the user will enter his OTP PIN and the sending of the Email is triggered. The user is denied the access.

Seconds step

In the second step the user authenticates with the OTP PIN and the OTP value he received via Email. The user is granted access.

Alternatively the user can authenticate with the transaction_id that was sent to him in the response during the first step and only the OTP value. The transaction_id assures that the user already presented the first factor (OTP PIN) successfully.

Configuration Parameters

You can configure the mail parameters for the Email Token centrally at Config -> Tokens -> Email.

Mail Server

The name or IP address of the mail server that is used to send emails.

Port

The port of the mail server.

Mail User

If the mail server requires authentication you need to enter a username. If no username is entered, no authentication is performed on the mail server.

Mail User Password

The password of the mail username to send emails.

Mail Sender Address

The mail address of the mail sender. This needs to correspond to the Mail User.

OTP validity time

This is the time in seconds, for how long the sent OTP value is valid. If a user tries to authenticate with the sent OTP value after this time, authentication will fail.

Use TLS

Whether the mail server should use TLS.

HOTP Token Config

HOTP Token configuration

SMS OTP Token

The SMS OTP token creates a OTP value and sends this OTP value to the mobile phone of the user. The SMS can be triggered by authenticating with only the OTP PIN:

First step

In the first step the user will enter his OTP PIN and the sending of the SMS is triggered. The user is denied the access.

Second step

In the second step the user authenticates with the OTP PIN and the OTP value he received via SMS. The user is granted access.

Alternatively the user can authenticate with the transaction_id that was sent to him in the response during the first step and only the OTP value. The transaction_id assures that the user already presented the first factor (OTP PIN) successfully.

A python SMS provider module defines how the SMS is sent. This can be done using an HTTP SMS Gateway. Most services like Clickatel or sendsms.de provide such a simple HTTP gateway. Another possibility is to send SMS via sipgate, which provides an XMLRPC API. The third possibility is to send the SMS via an SMTP gateway. The provider receives a specially designed email and sends the SMS accordingly. The last possibility to send SMS is to use an attached GSM modem.

Starting with version 2.13 the SMS configuration has been redesigned. You can now centrally define SMS gateways. These SMS gateways can be used for sending SMS OTP token but also for the event notifications. (See User Notification Handler Module)

For configuring SMS Gateways read SMS Gateway configuration. I this token configuration you can select on defined gateway to send SMS for authentication.

TiQR Token Config

TiQR Token configuration

TiQR Registration Server

You need at least enter the TiQR Registration Server. This is the URL of your privacyIDEA installation, that can be reached from the smartphone during enrollment. So your smartphone needs to be on the same LAN (WLAN) like the privacyIDEA server or the enrollment URL needs to be accessible from the internet.

You also need to specify the path, which is usually /ttype/tiqr.

During enrollment the parameter action=metadata and action=enrollment is added.

Note

We do not recommend putting the registration URL on the internet.

TiQR Authentication Server

This is the URL that is used during authentication. This can be another URL than the Registration Server. If it is left blank, the URL of the Registration Server is used.

During authentication the parameter operation=login is added.

TOTP Token Config

TOTP Token configuration

U2F Token Config

AppId

You need to configure the AppId of the privacyIDEA server. The AppId is define in the FIDO specification [1].

The AppId is the URL of your privacyIDEA and used to find or create the right key pair on the U2F device. The AppId must correspond the the URL that is used to call the privacyIDEA server.

Note

if you register a U2F device with an AppId https://privacyidea.example.com and try to authenticate at https://10.0.0.1, the U2F authentication will fail.

Note

The AppId must not contain any trailing slashes!

Facets

If specifying the AppId as the FQDN you will only be able to authenticate at the privacyIDEA server itself or at any application in a sub directory on the privacyIDEA server. This is OK, if you are running a SAML IdP on the same server.

But if you also want to use the U2F token with other applications, you need to specify the AppId like this:

https://privacyidea.example.com/pi-url/ttype/u2f

pi-url is the path, if you are running the privacyIDEA instance in a sub folder.

/ttype/u2f is the endpoint that returns a trusted facets list. Trusted facets are other hosts in the domain example.com. You need to define a policy that contains a list of the other hosts (u2f_facets).

For more information on AppId and trusted facets see [1].

For further details and for information how to add U2F to your application you can see the code documentation at U2F Token.

Workflow

You can use a U2F token on privacyIDEA and other hosts in the same Domain. To do so you need to do the following steps:

1. Configure the AppId to reflect your privacyIDEA server:

   https://pi.your-network.com/ttype/u2f

   Add the path /ttype/u2f is crucial. Otherwise privacyIDEA will not return the trusted facets.

2. Define a policy with the list of trusted facets. (see u2f_facets). Add the FQDNs of the hosts to the policy:

   saml.your-network.com otherapp.your-network.com vpn.your-network.com

   Note

   The privacyIDEA plugin for simpleSAMLphp supports U2F with privacyIDEA starting with version 2.8.

3. Now register a U2F token on https://pi.your-network.com. Due to the trusted facets you will also be able to use this U2F token on the other hosts.

4. Now got to https://saml.your-network.com and you will be able to authenticate with the very U2F token without any further registering.

Footnotes

[1]	(1, 2) https://fidoalliance.org/specs/fido-u2f-v1.0-nfc-bt-amendment-20150514/fido-appid-and-facets.html

Yubico Cloud mode

The Yubico Cloud mode sends the One Time Password emitted by the yubikey to the Yubico Cloud service or another (possibly self hosted) validation server.

Configure the Yubico Cloud mode

To contact the Yubico Cloud service you need to get an API key and a Client ID from Yubico and enter these here in the config dialog. In that case you can leave the Yubico URL blank and privacyidea will use the Yubico servers.

You can use another validation host, e.g. a self hosted validation server. If you use privacyidea token type yubikey, you can use the URL https://<privacyideaserver>/ttype/yubikey, other validation servers might use https://<validationserver>/wsapi/2.0/verify. You’ll get the Client ID and API key from the configuration of your validation server.

You can get your own API key at [1].

[1]	https://upgrade.yubico.com/getapikey/.

Yubikey AES mode

The Yubico AES mode uses the same kind of token as the Yubico Cloud service, but validates the OTP in your local privacyidea server. So the secrets stay local to your system and are not stored in Yubico’s Cloud service.

Configure the Yubikey AES mode

You can have more than one Client with a Client ID connect to your server. The Client ID starts with yubikey.apiid. and is followed by the API ID, which you’ll need to configure your clients. With create new API key you generate a new API for that specific Client ID. The API key is used to sign the validation request sent to the server and the server signs the answer too. That way tampering or MITM attacks might be detected. It is possible to validate token without the API key, but then the request and answer can’t be verify against the key. It is useful to use HTTPS for your validation requests, but this is another kind of protection.

OTP validation can either use the privacyidea API /validate/check or the Yubikey validation protocol /ttype/yubikey or - if enabled in your webserver configuration - /wsapi/2.0/verify.

CA Connectors

You can use privacyIDEA to enroll certificates and assign certificates to users.

You can define connections to Certifacte Authorities, that are used when enrolling certificates.

A local CA definition

When you enroll a Token of type certificate the Certificate Signing Request gets signed by one of the CAs attached to privacyIDEA by the CA connectors.

The first CA connector that ships with privacyIDEA is a connector to a local openSSL based Certificate Authority as shown in figure A local CA definition.

When enrolling a certificate token you can choose, which CA should sign the certificate request.

Enrolling a certificate token

Local CA Connector

The local CA connector calls a local openssl configuration.

Starting with privacyIDEA version 2.12 an example openssl.cnf is provided in /etc/privacyidea/CA/openssl.cnf.

Note

This configuration and also this description is ment to be as an example. When setting up a productive CA, you should ask a PKI consultant for assistance.

Manual Setup

1. Modify the parameters in the file /etc/privacyidea/CA/openssl.cnf according to your needs.

2. Create your CA certificate:

   openssl req -days 1500 -new -x509 -keyout /etc/privacyidea/CA/ca.key \
               -out /etc/privacyidea/CA/ca.crt \
               -config /etc/privacyidea/CA/openssl.cnf
   
   chmod 0600 /etc/privacyidea/CA/ca.key
   touch /etc/privacyidea/CA/index.txt
   echo 01 > /etc/privacyidea/CA/serial
   chown -R privacyidea /etc/privacyIDEA/CA
   

3. Now set up a local CA connector within privacyIDEA with the directory /etc/privacyidea/CA and the files accordingly.

Easy Setup

Starting with privacyIDEA version 2.18 it gets easier to setup local CAs.

You can use the pi-manage tool to setup a new CA like this:

pi-manage ca create myCA

This will ask you for all necessary parameters for the CA and then automatically

1. Create the files for this new CA and
2. Create the CA connector in privacyIDEA.

Management

There are different ways to enroll a certificate token. See Certificates.

When an administrator revokes a certificate token, the certificate is revoked and a CRL is created.

Note

privacyIDEA does not create the CRL regularly. The CRL usually has a validity period of 30 days. I.e. you need to create the CRL on a regular basis. You can use openssl to do so or the pi-manage command.

Starting with version 2.18 the pi-manage command has an additional sub-command ca:

pi-manage ca list

lists all configured CA connectors. You can use the -v switch to get more information.

You can create a new CRL with the command:

pi-manage ca create_crl <CA name>

This command will check the overlap period and only create a new CRL if it is necessary. If you want to force the creation of the CRL, you can use the switch -f.

For more information on pi-manage see The pi-manage Script.

Templates

The local CA supports a kind of certificate templates. These “templates” are predefined combinations of extensions and validity days, as they are passed to openssl via the parameters -extensions and -days.

This way the administrator can define certificate templates with certain X.509 extensions like keyUsage, extendedKeyUsage, CDPs or AIAs and certificate validity periods.

The extensions are defined in YAML file and the location of this file is added to the CA connector definition.

The file can look like this, defining three templates “user”, “webserver” and “template3”:

user:

days: 365 extensions: “user”

webserver:

days: 750 extensions: “server”

template3:

days: 10 extensions: “user”

SMTP server configuration

Starting with privacyIDEA 2.10 you can define SMTP server configurations. SMTP server endpoints.

An SMTP server configuration contains the

• server as FQDN or IP address,
• the port,
• the sender email address,
• a username and password in case of authentication and
• a TLS flag.

Each SMTP server configuration is address via a unique identifier. You can then use such a configuration for Email or SMS token, for PIN handling or for User registration.

Under Config->Sytem->SMTP servers you can get a list of all configured SMTP servers, create new server definitions and delete them.

The list of SMTP servers.

Using the unique identifier like themis you can use this SMTP server definition in e.g. a policy for user registraion.

Edit an existing SMTP server definition.

In the edit dialog you can enter all necessary attributes to talk to the SMTP server. You can also send a test email, to verify if your settings are correct.

SMS Gateway configuration

You can centrally define SMS gateways that can be used to send SMS with the SMS token (SMS OTP Token) or to use the SMS gateway for sending notifications.

There are different providers (gateways) to deliver SMS.

HTTP provider

The HTTP provider can be used for any SMS gateway that provides a simple HTTP POST or GET request. This is the most commonly used provider. Each provider type defines its own set of parameters.

The following parameters can be used. These are parameters, that define the behaviour of the SMS Gateway definition.

URL

This is the URL for the gateway.

HTTP_METHOD

Can be GET or POST.

USERNAME and PASSWORD

These are the username and the password if the HTTP request requires basic authentication.

RETURN_SUCCESS

You can either use RETURN_SUCCESS or RETURN_FAIL. If the text of RETURN_SUCCESS is found in the HTTP response of the gateway privacyIDEA assumes that the SMS was sent successfully.

RETURN_FAIL

If the text of RETURN_FAIL is found in the HTTP response of the gateway privacyIDEA assumes that the SMS could not be sent and an error occurred.

PROXY

You can specify a proxy to connect to the HTTP gateway.

PARAMETER

This can contain a dictionary of arbitrary fixed additional parameters. Usually this would also contain an ID or a password to identify you as a sender.

CHECK_SSL

If the URL is secured via TLS (HTTPS), you can select, if the certificate should be verified or not.

TIMEOUT

The timeout for contacting the API and receiving a response.

Options

You can define additional options. These are sent as parameters in the GET or POST request.

Note

The fixed parameters and the options can not have the same name! If you need an options, that has the same name as a parameter, you must not fill in the corresponding parameter.

Note

You can use the tags {phone} and {otp} to specify the mobile number and the otp value.

Examples

Clickatell

In case of the Clickatell provider the configuration will look like this:

• URL: http://api.clickatell.com/http/sendmsg
• HTTP_METHOD: GET
• RETURN_SUCCESS: ID

Set the additional options to be passed as HTTP GET parameters:

• user: YOU
• password: your password
• api_id: you API ID
• text: “Your OTP value is {otp}”
• to: {phone}

This will consturct an HTTP GET request like this:

http://api.clickatell.com/http/sendmsg?user=YOU&password=YOU&\
     api_id=YOUR API ID&text=....&to=....

where text and to will contain the OTP value and the mobile phone number. privacyIDEA will assume a successful sent SMS if the response contains the text “ID”.

GTX-Messaging

GTX-Messaging is an SMS Gateway located in Germany.

The configuration looks like this (see [2]):

• URL: https://http.gtx-messaging.net/smsc.php
• HTTP_METHOD: GET
• CHECK_SSL: yes
• RETURN_SUCCESS: 200 OK

You need to set the additional options:

• user: <your account>
• pass: <the account password>
• to: {phone}
• text: Your OTP value is {otp}.

Note

The user and pass are not the credentials you use to login. You can find the required credentials for sending SMS in your GTX messaging account when viewing the details of your routing account.

Twilio

You can also use the Twilio service for sending SMS. [1].

• URL: https://api.twilio.com/2010-04-01/Accounts/B...8/Messages
• HTTP_METHOD: POST

For basic authentication you need:

• USERNAME: your accountSid
• PASSWORD: your password

Set the additional options as POST parameters:

• From: your Twilio phone number
• Body: {otp}
• To: {phone}

Sipgate provider

The sipgate provider connects to https://samurai.sipgate.net/RPC2 and takes only two arguments USERNAME and PASSWORD.

Parameters:

USERNAME

The sipgate username.

PASSWORD

The sipgate password.

PROXY

You can specify a proxy to connect to the HTTP gateway.

It takes not options.

If you activate debug log level you will see the submitted SMS and the response content from the Sipgate gateway.

SMTP provider

The SMTP provider sends an email to an email gateway. This is a specified, fixed mail address.

The mail should contain the phone number and the OTP value. The email gateway will send the OTP via SMS to the given phone number.

SMTPIDENTIFIED

Here you can select on of your centrally defined SMTP servers.

MAILTO

This is the address where the email with the OTP value will be sent. Usually this is a fixed email address provided by your SMTP Gateway provider. But you can also use the tags {phone} and {otp} to replace the phone number or the one time password.

SUBJECT

This is the subject of the email to be sent. You can use the tags {phone} and {otp} to replace the phone number or the one time password.

BODY

This is the body of the email. You can use this to explain the user, what he should do with this email. You can use the tags {phone} and {otp} to replace the phone number or the one time password.

The default SUBJECT is set to {phone} and the default BODY to {otp}. You may change the SUBJECT and the BODY accordingly.

SMPP Provider

The SMPP provider was added in privacyIDEA 2.22. It uses an SMS Center via the SMPP protocol to deliver SMS to the users.

You need to specify the SMSC_HOST and SMSC_PORT to talk to the SMS center. privacyIDEA need to authenticate against the SMS center. For this you can add the parameters SYSTEM_ID and PASSWORD. The parameter S_ADDR is the sender’s number, shown to the users receiving an SMS. For the other parameters contact your SMS center operator.

[1]	https://www.twilio.com/docs/api/rest/sending-messages

[2]	https://www.gtx-messaging.com/de/api-docs/http/

privacyIDEA setup tool

privacyIDEA comes with a graphical setup tool to manage your token administrators and RADIUS clients. Thus you will get a kind of appliance experience. To install all necessary components read appliance.

To configure the system, login as the user root on your machine and run the command:

privacyidea-setup

This will bring you to this start screen.

Start screen of the appliance setup tool.

You can configure privacyidea settings, the log level, administrators, encryption key and much more. You can configure the webserver settings and RADIUS clients.

Configure privacyidea

You can create new token administrators, delete them and change their passwords.

In the FreeRADIUS settings you can create and delete RADIUS clients.

All changes done in this setup tool are directly read from and written to the corresponding configuration files. The setup tool parses the original nginx and freeradius configuration files. So there is no additional place where this data is kept.

Note

You can also edit the clients.conf and other configuration files manually. The setup tool will also read those manual changes!

Backup and Restore

Starting with version 1.5 the setup tool also supports backup and restore. Backups are written to the directory /var/lib/privacyidea/backup.

The backup contains all privacyIDEA configuration, the contents of the directory /etc/privacyidea, the encryption key, the configured administrators, the complete token database (MySQL) and Audit log. Furthermore if you are running FreeRADIUS the backup also contains the /etc/freeradius/clients.conf file.

Schedulded backup

At the configuration point Configure Backup you can define times when a scheduled backup should be performed. This information is written to the file /etc/crontab.

Scheduled backup

You can enter minutes, hours, day of month, month and day of week. If the entry should be valid for each e.g. month or hour, you need to enter a ‘*’.

In this example the 10 17 * * * (minute=10, hour=17) means to perform a backup each day and each month at 17:10 (5:10pm).

The example 1 10 1 * * (minute=1, hour=10, day of month=1) means to perform a backup on the first day of each month at 10:01 am.

Thus you could also perform backups only once a week at the weekend.

Immediate backup

If you want to run a backup right now you can choose the entry Backup now.

Restore

The entry View Backups will list all the backups available.

All available backups

You can select a backup and you are asked if you want to restore the data.

Warning

Existing data is overwritten and will be lost.

Components

Starting with privacyIDEA 2.15 you can see privacyIDEA components in the Web UI. privacyIDEA collects authenticating clients with their User Agent. Usually this is a type like PAM, FreeRADIUS, OTRS, Wordpress... This overview helps you to understand your network and keep track which clients are connected to your network.

components

Tokenview

The administrator can see all the tokens of all realms he is allowed to manage in the tokenview. Each token can be located in several realms and be assigned to one user. The administrator can see all the details of the token.

Token View

The administrator can click on one token, to show more details of this token and to perform actions on this token.

Token Details

The Token Details give you more information about the token and also let the administrator perform specific tasks for this token.

Token Detail

At the bottom you see the assigned user. You can click on the username and change to the User Details.

Lost token

When a user has lost a token, the administrator or the user can create a temporary password token for the user to login.

The administrator has to select the token that was lost and click the button Lost token. A new token of type PW is generated. The OTP PIN of the old token is automatically copied to the new token. Thus the administrator does not know the OTP PIN, while the user can use his old PIN.

A long password is displayed to the administrator and the administrator can read this password to the user. The user now can authenticate with his old OTP PIN and the long password.

The lost token is deactivated.

Get Serial

The administrator can enter a OTP value that was generated by an unknown token. Then the serial number for the corresponding token is search and displayed.

Note

Since OTP values for all matching tokens need to be calculated, this can be time consuming!

Token settings

You can change the following token settings.

MaxFail and FailCount

If the login fail counter reaches the MaxFail the user can not login with this token anymore. The Failcounter FailCount has to be reset to zero.

TokenDesc

The token description is also displayed in the tokenview. You can set a description to make it easier to identify a token.

CountWindow

The CountWindow is the look ahead window of event based tokens. If the user pressed the button on an event based token the counter in the token is increased. If the user does not use this otp value to authenticate, the server does not know, that the counter in the token was increased. This way the counter in the token can get out of sync with the server.

SyncWindow

If a token was out of sync (see CountWindow), then it needs to be synchronized. This is done by entering two consecutive OTP values. The server searches these two values within the next CountWindow (default 1000) values.

OtpLen

This is the length of the OTP value that is generated by the token. The password that is entered by the user is split according to this length. 6 or 8 characters are split as OTP value and the rest is used as static password (OTP PIN).

Hashlib

The HOTP algorithm can be used with SHA1 or SHA256.

Tokeninfo - Auth max

The administrator can set a value how often this token may be used for authentication. If the number of authentication try exceed this value, the token can not be used, until this Auth max value is increased.

Note

This way you could create tokens, that can be used only once.

Tokeninfo - Auth max success

The administrator can set a value how often this token may be used to successfully authenticate.

Tokeninfo - Valid start

A timestamp can be set. The token will only be usable for authentication after this start time.

Tokeninfo - Valid end

A timestamp can be set. The token can only be used before this end time.

Note

This way you can create temporary tokens for guests or short time or season employees.

Resync Token

The administrator can select one token and then enter two consecutive OTP values to resynchonize the token if it was out of sync.

set token realm

A token can be assigned to several realms. This is important if you have administrators for different realms. A realm administrator is only allowed to see tokens within his realms. He will not see tokens, that are not in his realm. So you can assign a token to realm A and realm B, thus the administrator A and the administrator B will be able to see the token.

get OTP

If the corresponding getOTP policy (Policies) is set, the administrator can get the OTP values of a token from the server without having the token with him.

Note

Of course this is a potential backdoor, since the administrator could login as the user/owner of this very token.

enroll

You can enroll a token either from the Token View or from the User Details. When enrolling a token from the User Details the token is directly assigned to the user.

If you enroll the token from the token view, you can select a user, to whom the token will be assigned.

When enrolling a token, you can select the token type and according to the token type other necessary information.

Token enrollment dialog

assign

This function is used to assign a token to a user. Select a realm and start typing a username to find the user, to whom the token should be assigned.

unassign

In the token details view you can unassign the token. After that, the token can be assigned to a new user.

enable

If a token is disabled, it can be enabled again.

disable

Tokens can be disabled. Disabled tokens still belong to the assigned user but those tokens can not be used to authenticate. Disabled tokens can be enabled again.

set PIN

You can set the OTP PIN or the mOTP PIN for tokens.

Reset Failcounter

If a used locked his token, since he entered wrong OTP values or wrong OTP PINs, the fail counter has reached the mail failcount. The administrator or help desk user can select those tokens and click the button reset failcounter to reset the fail counter to zero. The tokens can be used for authentication again.

delete

Deleting a token will remove the token from the database. The token information can not be recovered. But all events that occurred with this token still remain in the audit log.

Userview

The administrator can see all users in realms he is allowed to manage.

Note

Users are only visible, if the useridresolver is located within a realm. If you only define a useridresolver but no realm, you will not be able to see the users!

You can select one of the realms in the left drop down box. The administrator will only see the realms in the drop down box, that he is allowed to manage. (TODO) No migrated, yet.

User View. List all users in a realm.

The list shows the users from the select realm. The username, surname, given name, email and phone are filled according to the definition of the useridresolver.

Even if a realm contains several useridresolvers all users from all resolvers within this realm are displayed.

User Details

When clicking on a username, you can see the users details and perform several actions on the user.

User Details.

You see a list of the users tokens and change to the Token Details.

Enroll tokens

In the users details view you can enroll additional tokens to the user. In the enrollment dialog the user will be selected and you only need to choose what tokentype you wish to enroll for this user.

Assign tokens

You can assign a new, already existing token to the user. Just start typing the token serial number. The system will search for tokens, that are not assigned yet and present you a list to choose from.

View Audit Log

You can also click View user in Audit log which will take you to the Audit log with a filter on this very user, so that you will only see audit entries regarding this user.

Edit user

If the user is located in a resolver, that is marked as editable, the administrator will also see a button “Edit User”. To read more about this, see Manage Users.

Manage Users

Since version 2.4 privacyIDEA allows you to edit users in the configured resolvers. At the moment this is possible for SQL resolvers.

In the resolver definition you need to check the new checkbox Edit user store.

Users in SQL can be edited, when checking the checkbox.

In the Users Detail view, the administrator then can click the button “Edit” and modify the user data and also set a new password.

Edit the attributes of an existing user.

Note

The data of the user will be modified in the user store (database). Thus the users data, which will be returned by a resolver, is changed. If the resolver is contained in several realms these changes will reflect in all realms.

If you want to add a user, you can click on Add User in the User View.

Add a new user.

Users are contained in resolvers and added to resolvers. So you need to choose an existing resolver and not a realm. The user will be visible in all realms, the resolver is contained in.

Note

Of course you can set policies to allow or deny the administrator these rights.

Simple local users setup

You can setup a local users definition quite easily. Run:

pi-manage resolver create_internal test

This will create a database table “users_test” in your token database. And it will create a resolver “test” that refers to this database table.

Then you can add this resolver to realm:

pi-manage realm create internal_realm test

Which will create a realm “internal_realm” containing the resolver “test”. Now you can start adding users to this resolver as described above.

Note

This is an example of how to get started with users quite quickly. Of course you do not need to save the users table in the same database as the tokens. But in scenarios, where you do not have existing user stores or the user stores are managed by another department or are not accessible easily this may be sensible way.

Policies

Policies can be used to define the reaction and behaviour of the system.

Each policy defines the behaviour in a certain area, called scope. privacyIDEA knows the scopes:

Admin policies

Admin policies are used to regulate the actions that administrators are allowed to do. Technically admin policies control the use of the REST API Token endpoints, System endpoints, Realm endpoints and Resolver endpoints.

Admin policies are implemented as decorators in Policy Module and Policy Decorators.

The user in the admin policies refers to the name of the administrator.

Starting with privacyIDEA 2.4 admin policies can also store a field “admin realm”. This is used, if you define realms to be superuser realms. See The Config File for information how to do this. Read So what’s the thing with all the admins? for more information on the admin realms.

This way it is easy to define administrative rights for big groups of administrative users like help desk users in the IT department.

Admin scope provides and additional field ‘admin realm’.

All administrative actions also refer to the defined user realm. Meaning an administrator may have many rights in one user realm and only a few rights in another realm.

Creating a policy with scope:admin, admin-realm:helpdesk, user:frank, action:enable and realm:sales means that the administrator frank in the admin-realm helpdesk is allowed to enable tokens in the user-realm sales.

Note

As long as no admin policy is defined all administrators are allowed to do everything.

The following actions are available in the scope admin:

init

type: bool

There are init actions per token type. Thus you can create policy that allow an administrator to enroll SMS tokens but not to enroll HMAC tokens.

enable

type: bool

The enable action allows the administrator to activate disabled tokens.

disable

type: bool

Tokens can be enabled and disabled. Disabled tokens can not be used to authenticate. The disable action allows the administrator to disable tokens.

revoke

type: bool

Tokens can be revoked. Usually this means the token is disabled and locked. A locked token can not be modified anymore. It can only be deleted.

Certain token types like certificate may define special actions when revoking a token.

set

type: bool

Tokens can have additional token information, which can be viewed in the Token Details.

If the set action is defined, the administrator allowed to set those token information.

setpin

type: bool

If the setpin action is defined, the administrator is allowed to set the OTP PIN of a token.

enrollpin

type: bool

If the action enrollpin is defined, the administrator can set a token PIN during enrollment. If the action is not defined and the administrator tries to set a PIN during enrollment, this PIN is deleted from the request.

otp_pin_maxlength

type: integer

range: 0 - 31

This is the maximum allowed PIN length the admin is allowed to use when setting the OTP PIN.

Note

There can be token type specific policies like

spass_otp_pin_maxlength, spass_otp_pin_minlength and spass_otp_pin_contents. If suche a token specific policy exists, it takes priority of the common PIN policy.

otp_pin_minlength

type: integer

range: 0 - 31

This is the minimum required PIN the admin must use when setting the OTP PIN.

otp_pin_contents

type: string

contents: cns

This defines what characters an OTP PIN should contain when the admin sets it.

c are letters matching [a-zA-Z].

n are digits matching [0-9].

s are special characters matching [.:,;-_<>+*!/()=?$§%&#~^].

Example: The policy action otp_pin_contents=cn, otp_pin_minlength=8 would require the admin to choose OTP PINs that consist of letters and digits which have a minimum length of 8.

cn

test1234 and test12$$ would be valid OTP PINs. testABCD would not be a valid OTP PIN.

The logic of the otp_pin_contents can be enhanced and reversed using the characters + and -.

-cn would still mean, that the OTP PIN needs to contain letters and digits and it must not contain any other characters.

-cn (substraction)

test1234 would be a valid OTP PIN, but test12$$ and testABCS would not be valid OTP PINs. The later since it does not contain digits, the first (test12$$) since it does contain a special character ($), which it should not.

+cn (grouping)

combines the two required groups. I.e. the OTP PIN should contain characters from the sum of the two groups. test1234, test12$$, test and 1234 would all be valid OTP PINs.

resync

type: bool

If the resync action is defined, the administrator is allowed to resynchronize a token.

assign

type: bool

If the assign action is defined, the administrator is allowed to assign a token to a user. This is used for assigning an existing token to a user but also to enroll a new token to a user.

Without this action, the administrator can not create a connection (assignment) between a user and a token.

unassign

type: bool

If the unassign action is defined, the administrator is allowed to unassign tokens from a user. I.e. the administrator can remove the link between the token and the user. The token still continues to exist in the system.

import

type: bool

If the import action is defined, the administrator is allowed to import token seeds from a token file, thus creating many new token objects in the systems database.

remove

type: bool

If the remove action is defined, the administrator is allowed to delete a token from the system.

Note

If a token is removed, it can not be recovered.

Note

All audit entries of this token still exist in the audit log.

userlist

type: bool

If the userlist action is defined, the administrator is allowed to view the user list in a realm. An administrator might not be allowed to list the users, if he should only work with tokens, but not see all users at once.

Note

If an administrator has any right in a realm, the administrator is also allowed to view the token list.

checkstatus

type: bool

If the checkstatus action is defined, the administrator is allowed to check the status of open challenge requests.

manageToken

type: bool

If the manageToken action is defined, the administrator is allowed to manage the realms of a token.

A token may be located in multiple realms. This can be interesting if you have a pool of spare tokens and several realms but want to make the spare tokens available to several realm administrators. (Administrators, who have only rights in one realm)

Then all administrators can see these tokens and assign the tokens. But as soon as the token is assigned to a user in one realm, the administrator of another realm can not manage the token anymore.

getserial

type: bool

If the getserial action is defined, the administrator is allowed to calculate the token serial number for a given OTP value.

getrandom

type: bool

The getrandom action allows the administrator to retrieve random keys from the endpoint getrandom. This is an endpoint in System endpoints.

getrandom can be used by the client, if the client has no reliable random number generator. Creating API keys for the Yubico Validation Protocol uses this endpoint.

getchallenges

type: bool

This policy allows the administrator to retrieve a list of active challenges of a challenge response tokens. The administrator can view these challenges in the web UI.

losttoken

type: bool

If the losttoken action is defined, the administrator is allowed to perform the lost token process.

To only perform the lost token process the actions copytokenuser and copytokenpin are not necessary!

adduser

type: bool

If the adduser action is defined, the administrator is allowed to add users to a user store.

Note

The user store still must be defined as editable, otherwise no users can be added, edited or deleted.

updateuser

type: bool

If the updateuser action is defined, the administrator is allowed to edit users in the user store.

deleteuser

type: bool

If the deleteuser action is defined, the administrator is allowed to delete an existing user from the user store.

copytokenuser

type: bool

If the copytokenuser action is defined, the administrator is allowed to copy the user assignment of one token to another.

This functionality is also used during the lost token process. But you only need to define this action, if the administrator should be able to perform this task manually.

copytokenpin

type: bool

If the copytokenpin action is defined, the administrator is allowed to copy the OTP PIN from one token to another without knowing the PIN.

This functionality is also used during the lost token process. But you only need to define this action, if the administrator should be able to perform this task manually.

smtpserver_write

type: bool

To be able to define new SMTP server configuration or delete existing ones, the administrator needs this rights smtpserver_write.

eventhandling_write

type: bool

Allow the admiinstrator to configure Event Handler.

auditlog

type: bool

The administrators are allowed to view the audit log. If the policy contains a user realm, than the administrator is only allowed to see entries which contain this very user realm. A list of user realms may be defined.

To learn more about the audit log, see Audit.

auditlog_download

type: bool

The administrator is allowed to download the audit log.

Note

The download is not restricted to filters and audit age. Thus, if you want to avoid, that an administrator can see older logs, you need to disallow downloading the data. Otherwise he may download the audit log and look at older entries manually.

auditlog_age

type: string

This limits the maximum age of displayed audit entries. Older entries are not remove from the audit table but the administrator is simply not allowed to view older entries.

Can be something like 10m (10 minutes), 10h (10 hours) or 10d (ten days).

trigger_challenge

type: bool

If set the administrator is allowed to call the API /validate/triggerchallenge. This API can be used to send an OTP SMS to user without having specified the PIN of the SMS token.

The usual setup that one administrative account has only this single policy and is only used for triggering challenges.

New in version 2.17.

hotp_2step and totp_2step

type: string

This allows or forces the administrator to enroll a smartphone based token in two steps. In the second step the smartphone generates a part of the OTP secret, which the administrator needs to enter. (see Two Step Enrollment). Possible values are allow and force. This works in conjunction with the enrollment parameters {type}_2step_clientsize, {type}_2step_serversize, {type}_2step_difficulty.

Such a policy can also be set for the user. See hotp_2step and totp_2step.

New in version 2.21

User Policies

In the Web UI users can manage their own tokens. User can login to the Web UI with the username of their useridresolver. I.e. if a user is found in an LDAP resolver pointing to Active Directory the user needs to login with his domain password.

User policies are used to define, which actions users are allowed to perform.

The user policies also respect the client input, where you can enter a list of IP addresses and subnets (like 10.2.0.0/16).

Using the client parameter you can allow different actions in if the user either logs in from the internal network or remotely from the internet via the firewall.

Technically user policies control the use of the REST API Token endpoints and are checked using Policy Module and Policy Decorators.

Note

If no user policy is defined, the user has all actions available to him, to manage his tokens.

The following actions are available in the scope user:

enroll

type: bool

There are enroll actions per token type. Thus you can create policies that allow the user to enroll SMS tokens but not to enroll HMAC tokens.

assgin

type: bool

The user is allowed to assgin an existing token, that is located in his realm and that does not belong to any other user, by entering the serial number.

disable

type: bool

The user is allowed to disable his own tokens. Disabled tokens can not be used to authenticate.

enable

type: bool

The user is allowed to enable his own tokens.

delete

type: bool

The user is allowed to delete his own tokens from the database. Those tokens can not be recovered. Anyway, the audit log concerning these tokens remains.

unassign

type: bool

The user is allowed to drop his ownership of the token. The token does not belong to any user anymore and can be reassigned.

resync

type: bool

The user is allowed to resynchronize the token if it has got out of synchronization.

reset

type: bool

The user is allowed to reset the failcounter of the token.

setpin

type: bool

The user ist allowed to set the OTP PIN for his tokens.

enrollpin

type: bool

If the action enrollpin is defined, the user can set a token PIN during enrollment. If the action is not defined and the user tries to set a PIN during enrollment, this PIN is deleted from the request.

otp_pin_maxlength

type: integer

range: 0 - 31

This is the maximum allowed PIN length the user is allowed to use when setting the OTP PIN.

Note

There can be token type specific policies like

spass_otp_pin_maxlength, spass_otp_pin_minlength and spass_otp_pin_contents. If suche a token specific policy exists, it takes priority of the common PIN policy.

otp_pin_minlength

type: integer

range: 0 - 31

This is the minimum required PIN the user must use when setting the OTP PIN.

otp_pin_contents

type: string

contents: cns

This defines what characters an OTP PIN should contain when the user sets it.

c are letters matching [a-zA-Z].

n are digits matching [0-9].

s are special characters matching [.:,;-_<>+*!/()=?$§%&#~^].

Example: The policy action otp_pin_contents=cn, otp_pin_minlength=8 would require the user to choose OTP PINs that consist of letters and digits which have a minimum length of 8.

cn

test1234 and test12$$ would be valid OTP PINs. testABCD would not be a valid OTP PIN.

The logic of the otp_pin_contents can be enhanced and reversed using the characters + and -.

-cn would still mean, that the OTP PIN needs to contain letters and digits and it must not contain any other characters.

-cn (substraction)

test1234 would be a valid OTP PIN, but test12$$ and testABCS would not be valid OTP PINs. The later since it does not contain digits, the first (test12$$) since it does contain a special character ($), which it should not.

+cn (grouping)

combines the two required groups. I.e. the OTP PIN should contain characters from the sum of the two groups. test1234, test12$$, test and 1234 would all be valid OTP PINs.

auditlog

type: bool

This action allows the user to view and search the audit log for actions with his own tokens.

To learn more about the audit log, see Audit.

auditlog_age

type: string

This limits the maximum age of displayed audit entries. Older entries are not remove from the audit table but the user is simply not allowed to view older entries.

Can be something like 10m (10 minutes), 10h (10 hours) or 10d (ten days).

updateuser

type: bool

If the updateuser action is defined, the user is allowed to change his attributes in the user store.

Note

To be able to edit the attributes, the resolver must be defined as editable.

revoke

type: bool

Tokens can be revoked. Usually this means the token is disabled and locked. A locked token can not be modified anymore. It can only be deleted.

Certain token types like certificate may define special actions when revoking a token.

password_reset

type: bool

Introduced in version 2.10.

If the user is located in an editable user store, this policy can define, if the user is allowed to perform a password reset. During the password reset an email with a link to reset the password is sent to the user.

hotp_2step and totp_2step

type: string

This allows or forces the user to enroll a smartphone based token in two steps. In the second step the smartphone generates a part of the OTP secret, which the user needs to enter. (see Two Step Enrollment). Possible values are allow and force. This works in conjunction with the enrollment parameters {type}_2step_clientsize, {type}_2step_serversize, {type}_2step_difficulty.

Such a policy can also be set for the administrator. See hotp_2step and totp_2step.

New in version 2.21

Authentication policies

The scope authentication gives you more detailed possibilities to authenticate the user or to define what happens during authentication.

Technically the authentication policies apply to the REST API Validate endpoints and are checked using Policy Module and Policy Decorators.

The following actions are available in the scope authentication:

otppin

type: string

This action defines how the fixed password part during authentication should be validated. Each token has its own OTP PIN, but you can choose how the authentication should be processed:

otppin=tokenpin

This is the default behaviour. The user needs to pass the OTP PIN concatenated with the OTP value.

otppin=userstore

The user needs to pass the user store password concatenated with the OTP value. It does not matter if the OTP PIN is set or not. If the user is located in an Active Directory the user needs to pass his domain password together with the OTP value.

Note

The domain password is checked with an LDAP bind right at the moment of authentication. So if the user is locked or the password was changed authentication will fail.

otppin=none

The user does not have to pass any fixed password. Authentication is only done via the OTP value.

passthru

type: str

If the user has no token assigned, he will be authenticated against the userstore or against the given RADIUS configuration. I.e. the user needs to provide the LDAP- or SQL-password or valid credentials for the RADIUS server.

Note

This is a good way to do a smooth enrollment. Users having a token enrolled will have to use the token, users not having a token, yet, will be able to authenticate with their domain password.

It is also a way to do smooth migrations from other OTP systems. The authentication request of users without a token is forwarded to the specified RADIUS server.

Note

The passthru policy overrides the authorization policy for tokentype. I.e. a user may authenticate due to the passthru policy (since he has no token) although a tokentype policy is active!

Warning

If the user has the right to delete his tokens in selfservice portal, the user could delete all his tokens and then authenticate with his static password again.

passOnNoToken

type: bool

If the user has no token assigned an authentication request for this user will always be true.

Warning

Only use this if you know exactly what you are doing.

passOnNoUser

type: bool

If the user does not exist, the authentication request is successful.

Warning

Only use this if you know exactly what you are doing.

smstext

type: string

This is the text that is sent via SMS to the user trying to authenticate with an SMS token. You can use the tags <otp> and <serial>.

Starting with version 2.20 you can use the tag {challenge}. This will add the challenge data that was passed in the first authentication request in the challenge parameter. This could contain banking transaction data.

Default: <otp>

smsautosend

type: bool

A new OTP value will be sent via SMS if the user authenticated successfully with his SMS token. Thus the user does not have to trigger a new SMS when he wants to login again.

emailtext

type: string

This is the text that is sent via Email to be used with Email Token. This text should contain the OTP value. You can use the tags <otp> and <serial>.

Starting with version 2.20 you can use the tag {challenge}. This will add the challenge data that was passed in the first authentication request in the challenge parameter. This could contain banking transaction data.

Default: <otp>

You can also provide the filename to an email template. The filename must be prefixed with file: like file:/etc/privacyidea/emailtemplate.html. The template is an HTML file and contain the tags {otp} and {serial}.

Note

If a message text is supplied directly, the email is sent as plain text. If the email template is read from a file, a HTML-only email is sent instead.

emailsubject

type: string

This is the subject of the Email sent by the Email Token. You can use the tags <otp> and <serial>.

Default: Your OTP

emailautosend

type: bool

If set, a new OTP Email will be sent, when successfully authenticated with an Email Token.

mangle

type: string

The mangle policy can mangle the authentication request data before they are processed. I.e. the parameters user, pass and realm can be modified prior to authentication.

This is useful if either information needs to be stripped or added to such a parameter. To accomplish that, the mangle policy can do a regular expression search and replace using the keyword user, pass (password) and realm.

A valid action could look like this:

action: mangle=user/.*(.{4})/user\\1/

This would modify a username like “userwithalongname” to “username”, since it would use the last four characters of the given username (“name”) and prepend the fixed string “user”.

This way you can add, remove or modify the contents of the three parameters. For more information on the regular expressions see [1].

Note

You must escape the backslash as \\ to refer to the found substrings.

Example: A policy to remove whitespace characters from the realm name would look like this:

action: mangle=realm/\\s//

Example: If you want to authenticate the user only by the OTP value, no matter what OTP PIN he enters, a policy might look like this:

action: mangle=pass/.*(.{6})/\\1/

Example: If you want to strip a string from the front of a username, for example to have “admin_username” resolve to just “username”, it would look like this:

action: mangle=user/admin_(.*)/\\1/

challenge_response

type: string

This is a list of token types for which challenge response can be used during authentication. The list is separated by whitespaces like “hotp totp”.

Note

The TiQR token does not need this setting, since it always works with challenge response.

u2f_facets

type: string

This is a white space separated list of domain names, that are trusted to also use a U2F device that was registered with privacyIDEA.

You need to specify a list of FQDNs without the https scheme like:

“host1.example.com host2.exmaple.com firewall.example.com”

For more information on configuring U2F see U2F Token Config.

[1]	https://docs.python.org/2/library/re.html

reset_all_user_tokens

type: bool

If a user authenticates successfully all failcounter of all of his tokens will be reset. This can be important, if using empty PINs or otppin=None.

auth_cache

type: string

The Authentication Cache caches the credentials of a successful authentication and allows to use the same credentials - also with an OTP value - for the specified amount of time.

The time to cache the credentials can be specified like “4h”, “5m”, “2d” (hours, minutes days) or “4h/5m”. The notation 4h/5m means, that credentials are cached for 4 hours, but only may be used again, if every 5 minutes the authentication occurs. If the authentication with the same credentials would not occur within 5 minutes, the credentials can not be used anymore.

In future implementations the caching of the credentials could also be dependent on the clients IP address and the user agent.

Note

The AuthCache only works for user authentication, not for authentication with serials.

Authorization policies

The scope authorization provides means to define what should happen if a user proved his identity and authenticated successfully.

Authorization policies take the realm, the user and the client into account.

Technically the authorization policies apply to the Validate endpoints and are checked using Policy Module and Policy Decorators.

The following actions are available in the scope authorization:

tokentype

type: string

Users will only be authorized with this very tokentype. The string can hold a space separated list of case sensitive tokentypes. It should look like:

hotp totp spass

This is checked after the authentication request, so that a valid OTP value is wasted, so that it can not be used, even if the user was not authorized at this request

Note

Combining this with the client IP you can use this to allow remote access to sensitive areas only with one special token type while allowing access to less sensitive areas with other token types.

serial

type: string

Users will only be authorized with the serial number. The string can hold a regular expression as serial number.

This is checked after the authentication request, so that a valid OTP value is wasted, so that it can not be used, even if the user was not authorized at this request

Note

Combining this with the client IP you can use this to allow remote access to sensitive areas only with hardware tokens like the Yubikey, while allowing access to less secure areas also with a Google Authenticator.

tokeninfo

type: string

Users will only be authorized if the tokeninfo field of the token matches this regular expression.

This is checked after the authentication request, so that a valid OTP value can not be used anymore, even if authorization is forbidden.

A valid action could look like

action = key/regexp/

Example:

action = last_auth/^2018.*/

This would mean the tokeninfo field needs to start with “2018”.

setrealm

type: string

This policy is checked before the user authenticates. The realm of the user matching this policy will be set to the realm in this action.

Note

This can be used if the user can not pass his realm when authenticating at a certain client, but the realm needs to be available during authentication since the user is not located in the default realm.

no_detail_on_success

type: bool

Usually an authentication response returns additional information like the serial number of the token that was used to authenticate or the reason why the authentication request failed.

If this action is set and the user authenticated successfully this additional information will not be returned.

no_detail_on_fail

type: bool

Usually an authentication response returns additional information like the serial number of the token that was used to authenticate or the reason why the authentication request failed.

If this action is set and the user fails to authenticate this additional information will not be returned.

api_key_required

type: bool

This policy is checked before the user is validated.

You can create an API key, that needs to be passed to use the validate API. If an API key is required, but no key is passed, the authentication request will not be processed. This is used to avoid denial of service attacks by a rogue user sending arbitrary requests, which could result in the token of a user being locked.

You can also define a policy with certain IP addresses without issuing API keys. This would result in “blocking” those IP addresses from using the validate endpoint.

You can issue API keys like this:

pi-manage api createtoken -r validate

The API key (Authorization token) which is generated is valid for 365 days.

The authorization token has to be used as described in Authentication endpoints.

auth_max_success

type: string

Here you can specify how many successful authentication requests a user is allowed to perform during a given time. If this value is exceeded, the authentication attempt is canceled.

Specify the value like 2/5m meaning 2 successful authentication requests per 5 minutes. If during the last 5 minutes 2 successful authentications were performed the authentication request is discarded. The used OTP value is invalidated.

Allowed time specifiers are s (second), m (minute) and h (hour).

auth_max_fail

type: string

Here you can specify how many failed authentication requests a user is allowed to perform during a given time.

If this value is exceeded, authentication is not possible anymore. The user will have to wait.

If this policy is not defined, the normal behaviour of the failcounter applies. (see Reset Fail Counter)

Specify the value like 2/1m meaning 2 successful authentication requests per minute. If during the last 5 minutes 2 successful authentications were performed the authentication request is discarded. The used OTP value is invalidated.

Allowed time specifiers are s (second), m (minute) and h (hour).

last_auth

type: string

You can define if an authentication should fail, if the token was not successfully used for a certain time.

Specify a value like 12h, 123d or 2y to disallow authentication, if the token was not successfully used for 12 hours, 123 days or 2 years.

The date of the last successful authentication is store in the tokeninfo field of a token and denoted in UTC.

u2f_req

type: string

Only the specified U2F devices are authorized to authenticate. The administrator can specify the action like this:

u2f_req=subject/.*Yubico.*/

The the key word can be “subject”, “issuer” or “serial”. Followed by a regular expression. During registration of the U2F device the information from the attestation certificate is stored in the tokeninfo. Only if the regexp matches this value, the authentication with such U2F device is authorized.

add_user_in_response

type: bool

In case of a successful authentication additional user information is added to the response. A dictionary containing user information is added in detail->user.

add_resolver_in_response

type: bool

In case of a successful authentication the resolver and realm of the user are added to the response. The names are added in detail->user-resolver and detail->user-realm.

Enrollment policies

The scope enrollment defines what happens during enrollment either by an administrator or during the user self enrollment.

Enrollment policies take the realms, the client (see Policies) and the user settings into account.

Technically enrollment policies control the use of the REST API Token endpoints and specially the init and assign-methods.

Technically the decorators in API Policies are used.

The following actions are available in the scope enrollment:

max_token_per_realm

type: int

This is the maximum allowed number of tokens in the specified realm.

Note

If you have several realms with realm admins and you imported a pool of hardware tokens you can thus limit the consumed hardware tokens per realm.

max_token_per_user

type: int

Limit the maximum number of tokens per user in this realm.

Note

If you do not set this action, a user may have unlimited tokens assigned.

tokenissuer

type: string

This sets the issuer label for a newly enrolled Google Authenticator. This policy takes a fixed string, to add additional information about the issuer of the soft token.

Starting with version 2.20 you can use the tags {user}, {realm}, {serial} and as new tags {givenname} and {surname} in the field issuer.

Note

A good idea is to set this to the instance name of your privacyIDEA installation or the name of your company.

tokenlabel

type: string

This sets the label for a newly enrolled Google Authenticator. Possible tags to be replaces are <u> for user, <r> for realm an <s> for the serial number.

The default behaviour is to use the serial number.

Note

This is useful to identify the token in the Authenticator App.

Note

Starting with version 2.19 the usage of <u>, <s> and <r> is deprecated. Instead you should use {user}, {realm}, {serial} and as new tags {givenname} and {surname}.

Warning

If you are only using <u> or {user} as tokenlabel and you enroll the token without a user, this will result in an invalid QR code, since it will have an empty label. You should rather use a label like “{user}@{realm}”, which would result in “@”.

autoassignment

type: string

allowed values: any_pin, userstore

Users can assign a token just by using this token. The user can take a token from a pool of unassigned tokens. When this policy is set, and the user has no token assigned, autoassignment will be done: The user authenticates with a new PIN or his userstore password and an OTP value from the token. If the OTP value is correct the token gets assigned to the user and the given PIN is set as the OTP PIN.

Note

Requirements are:

1. The user must have no other tokens assigned.
2. The token must be not assigned to any user.
3. The token must be located in the realm of the authenticating user.
4. (The user needs to enter the correct userstore password)

Warning

If you set the policy to any_pin the token will be assigned to the user no matter what pin he enters. In this case assigning the token is only a one-factor-authentication: the possession of the token.

otp_pin_random

type: int

Generates a random OTP PIN of the given length during enrollment. Thus the user is forced to set a certain OTP PIN.

Note

To use the random PIN, you also need to define a pinhandling policy.

pinhandling

type: string

If the otp_pin_random policy is defined, you can use this policy to define, what should happen with the random pin. The action value take the class of a PinHandler like privacyidea.lib.pinhandling.base.PinHandler. The base PinHandler just logs the PIN to the log file. You can add classes to send the PIN via EMail or print it in a letter.

For more information see the base class PinHandler.

change_pin_on_first_use

type: bool

If the administrator enrolls a token or resets a PIN of a token, then the PIN of this token is marked to be changed on the first (or next) use. When the user authenticates with the old PIN, the user is authenticated successfully. But the detail-response contains the keys “next_pin_change” and “pin_change”. If “pin_change” is True the authenticating application must trigger the change of the PIN using the API /token/setpin. See Token endpoints.

Note

If the application does not honour the “pin_change” attribute, then the user can still authenticate with his old PIN.

change_pin_every

type: string

This policy requires the user to change the PIN of his token on a regular basis. Enter a value follewed by “d”, e.g. change the PIN every 180 days will be “180d”.

The date, when the PIN needs to be changed, is returned in the API response of /validate/check. For more information see change_pin_on_first_use. To specifiy the contents of the PIN see User Policies.

otp_pin_encrypt

type: bool

If set the OTP PIN of a token will be encrypted. The default behaviour is to hash the OTP PIN, which is safer.

lostTokenPWLen

type: int

This is the length of the generated password for the lost token process.

lostTokenPWContents

type: string

This is the contents that a generated password for the lost token process should have. You can use

• c: for lowercase letters
• n: for digits
• s: for special characters (!#$%&()*+,-./:;<=>?@[]^_)
• C: for uppercase letters

Example:

The action lostTokenPWLen=10, lostTokenPWContents=Cns could generate a password like AC#!49MK)).

lostTokenValid

type: int

This is how many days the replacement token for the lost token should be valid. After this many days the replacement can not be used anymore.

yubikey_access_code

type: string

This is a 12 character long access code in hex format to be used to initialize yubikeys. If no access code is set, yubikeys can be re-initialized by everybody. You can choose a company wide access code, so that Yubikeys can only be re-initialized by your own system.

You can add two access codes separated by a colon to change from one access code to the other.

313233343536:414243444546

papertoken_count

type: int

This is a specific action of the paper token. Here the administrator can define how many OTP values should be printed on the paper token.

u2f_req

type: string

Only the specified U2F devices are allowed to be registered. The action can be specified like this:

u2f_req=subject/.*Yubico.*/

The the key word can be “subject”, “issuer” or “serial”. Followed by a regular expression. During registration of the U2F device the information is fetched from the attestation certificate. Only if the attribute in the attestation certificate matches accordingly the token can be registered.

u2f_no_verify_certificate

type: bool

By default the validity period of the attestation certificate of a U2F device gets verified during the registration process. If you do not want to verify the validity period, you can check this action.

{type}_2step_clientsize, {type}_2step_serversize, {type}_2step_difficulty

type: string

These are token type specific parameters. They control the key generation during the 2step token enrollment (see Two Step Enrollment).

The serversize is the optional size (in bytes) of the server’s key part. The clientsize is the size (in bytes) of the smartphone’s key part. The difficulty is a parameter for the key generation. In the implementation in version 2.21 PBKDF2 is used. In this case the difficulty specifies the number of rounds.

This is new in version 2.21

WebUI Policies

login_mode

type: string

allowed values: “userstore”, “privacyIDEA”, “disable”

If set to userstore (default), users and administrators need to authenticate with the password of their userstore, being an LDAP service or an SQL database.

If this action is set to login_mode=privacyIDEA, the users and administrators need to authenticate against privacyIDEA when logging into the WebUI. I.e. they can not login with their domain password anymore but need to authenticate with one of their tokens.

If set to login_mode=disable the users and administrators of the specified realms can not login to the UI anymore.

Warning

If you set this action and the user deletes or disables all his tokens, he will not be able to login anymore.

Note

Administrators defined in the database using the pi-manage command can still login with their normal passwords.

Note

A sensible way to use this, is to combine this action in a policy with the client parameter: requiring the users to login to the Web UI remotely from the internet with OTP but still login from within the LAN with the domain password.

Note

Another sensible way to use this policy is to disable the login to the web UI either for certain IP addresses (client) or for users in certain realms.

remote_user

type: string

This policy defines, if the login to the privacyIDEA using the web servers integrated authentication (like basic authentication or digest authentication) should be allowed.

Possible values are “disable” and “allowed”.

Note

The policy is evaluated before the user is logged in. At this point in time there is no realm known, so a policy to allow remote_user must not select any realm.

Note

The policy login_mode and remote_user work independent of each other. I.e. you can disable login_mode and allow remote_user.

You can use this policy to enable Single-Sign-On and integration into Kerberos or Active Directory. Add the following template into you apache configuration in /etc/apache2/sites-available/privacyidea.conf:

<Directory />
        # For Apache 2.4 you need to set this:
        # Require all granted
        Options FollowSymLinks
        AllowOverride None

        SSLRequireSSL
        AuthType Kerberos
        AuthName "Kerberos Login"
        KrbMethodNegotiate On
        KrbMethodK5Passwd On
        KrbAuthRealms YOUR-REALM
        Krb5KeyTab /etc/apache2/http.keytab
        KrbServiceName HTTP
        KrbSaveCredentials On
        <RequireAny>
            # Either we need a URL with no authentication or we need a valid user
            <RequireAny>
                # Any of these URL do NOT need a basic authentication
                Require expr %{REQUEST_URI} =~ m#^/validate#
                Require expr %{REQUEST_URI} =~ m#^/ttype#
            </RequireAny>
            Require valid-user
        </RequireAny>
</Directory>

logout_time

type: int

Set the timeout, after which a user in th WebUI will be logged out. The default timeout is 120 seconds.

Being a policy this time can be set based on clients, realms and users.

token_page_size

type: int

By default 15 tokens are displayed on one page in the token view. On big screens you might want to display more tokens. Thus you can define in this policy how many tokens should be displayed.

user_page_size

type: int

By default 15 users are displayed on one page in the user view. On big screens you might want to display more users. Thus you can define in this policy how many users should be displayed.

policy_template_url

type: str

Here you can define a URL from where the policies should be fetched. The default URL is a Github repository [#defaulturl]_.

Note

When setting a template_url policy the modified URL will only get active after the user has logged out and in again.

[1]	https://github.com/privacyidea/policy-templates/.

default_tokentype

type: str

You can define which is the default tokentype when enrolling a new token in the Web UI. This is the token, which will be selected, when entering the enrollment dialog.

tokenwizard

type: bool

If this policy is set and the user has no token, then the user will only see an easy token wizard to enroll his first token. If the user has enrolled his first token and he logs in to the web UI, he will see the normal view.

The user will enroll a token defined in default_tokentype.

Other sensible policies to combine are in User Policies the OTP length, the TOTP timestep and the HASH-lib.

You can add a prologue and epilog to the enrollment wizard in the greeting and after the token is enrolled and e.g. the QR code is displayed.

Create the files

• static/customize/views/includes/token.enroll.pre.top.html
• static/customize/views/includes/token.enroll.pre.bottom.html
• static/customize/views/includes/token.enroll.post.top.html
• static/customize/views/includes/token.enroll.post.bottom.html

to display the contents in the first step (pre) or in the second step (post).

Note

You can change the directory static/customize to a URL that fits your needs the best by defining a variable PI_CUSTOMIZATION in the file pi.cfg. This way you can put all modifications in one place apart from the original code.

realm_dropdown

type: str

If this policy is activated the web UI will display a realm dropdown box. Of course this policy can not filter for users or realms, since the user is not known at this moment.

The type of this action was changed to “string” in version 2.16. You can set a space separated list of realm names. Only these realmnames are displayed in the dropdown box.

Note

The realm names in the policy are not checked, if they realy exist!

search_on_enter

type: bool

The searching in the user list is performed as live search. Each time a key is pressed, the new substring is searched in the user store.

Sometimes this can be too time consuming. You can use this policy to change the bahaviour that the administrator needs to press enter to trigger the search.

(Since privacyIDEA 2.17)

custom_baseline

type: str

The administrator can replace the file templates/baseline.html with another template. This way he can change the links to e.g. internal documentation or ticketing systems. The new file could be called mytemplates/mybase.html.

This will only work with a valid subscription of privacyIDEA Enterprise Edition.

Note

This policy is evaluated before login. So any realm or user setting will have no effect. But you can specify different baselines for different client IP addresses.

(Since privacyIDEA 2.21)

custom_menu

type: str

The administrator can replace the file templates/menu.html with another template. This way he can change the links to e.g. internal documentation or ticketing systems. The new file could be called mytemplates/mymenu.html.

This will only work with a valid subscription of privacyIDEA Enterprise Edition.

Note

This policy is evaluated before login. So any realm or user setting will have no effect. But you can specify different menus for different client IP addresses.

(Since privacyIDEA 2.21)

Gettoken policies

The scope gettoken defines the maximum number of OTP values that may be retrieved from an OTP token by an administrator.

The user attribute may hold a list of administrators.

Technically the gettoken policies control the use of the gettoken_controller.

The following actions are available in the scope gettoken:

max_count_dpw

type: int

This is the maximum number of OTP values that are allowed to be retrieved from a DPW token.

Note

Issuing only one OTP value per day, this means that this is the number of days, this OTP list can be used.

max_count_hotp

type: int

This is the maximum number of OTP values that are allowed to be retrieved from an HOTP (HMAC) token.

Note

As hotp values only expire, when they are used, you can use this to create an OTP list, that can be used from the first to the last OTP value.

max_count_totp

type: int

This is the maximum number of OTP balues that are allowed to be retrieved from a TOTP token.

Note

As the default TOTP token generates a new OTP value all 30 seconds, retrieving 100 OTP values will only give you OTP values, that are usable for 50 minutes.

Register Policy

User registration

Starting with privacyIDEA 2.10 users are allowed to register with privacyIDEA. I.e. a user that does not exist in a given realm and resolver can create a new account.

Note

Registering new users is only possible, if there is a writeable resolver and if the necessary policy in the scope register is defined. For editable UserIdResolvers see UserIdResolvers.

If a register policy is defined, the login window of the Web UI gets a new link “Register”.

Next to the login button is a new link ‘register’, so that new users are able to register.

A user who clicks the link to register a new account gets this registration dialog:

Registration form

During registration the user is also enrolled Registration token. This registration code is sent to the user via a notification email.

Note

Thus - using the right policies in scope webui and authentication - the user could login with the password he set during registration an the registration code he received via email.

Policy settings

In the scope register several settings define the behaviour of the registration process.

Creating a new registration policy

realm

type: string

This is the realm, in which a new user will be registered. If this realm is not specified, the user will be registered in the default realm.

resolver

type: string

This is the resolver, in which the new user will be registered. If this resolver is not specified, registration is not possible!

Note

This resolver must be an editable resolver, otherwise the user can not be created in this resolver.

smtpconfig

type: string

This is the unique identifier of the SMTP server configuration. This SMTP server is used to send the notification email with the registration code during the registration process.

Note

If there is no smtpconfig or set to a wrong identifier, the user will get no notification email.

requiredemail

type: string

This is a regular expression according to [1].

Only email addresses matching this regular expression are allowed to register.

Example: If you want to authenticate the user only by the OTP value, no matter what OTP PIN he enters, a policy might look like this:

action: requiredemail=/.*@mydomain\..*/

This will allow all email addresses from the domains mydomain.com, mydomain.net etc...

[1]	https://docs.python.org/2/library/re.html

You can define as many policies as you wish to. The logic of the policies in the scopes is additive.

Policy Definition

Starting with privacyIDEA 2.5 you can use policy templates to ease the setup.

Policy Templates

Policy templates are defined in a Github repository which can be changed using a WebUI policy policy_template_url.

The policy templates are json files, which can contain common settings, that can be used to start your own polcies. When creating a new policy, you can select an existing policy template as a starting point.

You may also fork the github repository and commit pull request to improve the policy templates. Or you may fork the github repositry and use your own policy template URL for your policy templates.

A policy templates looks like this:

{
 "name": "template_name1",
 "scope": "enrollment",
 "action": {
            "tokenlabel": "<u>@<r>/<s>",
            "autoassignment": true
           }
}

realms, resolver and clients are not used in the templates.

A template must be referenced in a special index.json file:

{
  "template_name1": "description1",
  "template_name2": "description2"
}

where the key is the name of the template file and the value is a description displayed in the WebUI.

Each policy can contain the following attributes:

policy name

A unique name of the policy. The name is the identifier of the policy. If you create a new policy with the same name, the policy is overwritten.

scope

The scope of the policy as described above.

action

This is the important part of the policy. Each scope provides its own set of actions. An action describes that something is allowed or that some behaviour is configured. A policy can contain several actions. Actions can be of type boolean, string or integer. Boolean actions are enabled by just adding this action - like scope=user:action=disable, which allows the user to disable his own tokens. string and integer actions require an additional value - like scope=authentication:action='otppin=userstore'.

user

This is the user, for whom this policy is valid. Depending on the scope the user is either an administrator or a normal authenticating user.

If this field is left blank, this policy is valid for all users.

resolver

This policy will be valid for all users in this resolver.

If this field is left blank, this policy is valid for all resolvers.

Note

Starting with version 2.17 you can use the parameter check_all_resolvers. This is Check all possible resolvers of a user to match the resolver in this policy in the Web UI.

Assume a user user@realm1 is contained in resolver1 and resolver2 in the realm realm1, where resolver1 is the resolver with the highest priority. If the user authenticates as user@realm1, only policies for resolver1 will match, since the user is identified as user.resolver1@realm1.

If you also want to match a policy with resolver=resolver2, you need to select Check all possible resolvers in this policy. Thus this policy will match for all users, which are als contained in resolver2 as a secondary resolver.

realm

This is the realm, for which this policy is valid.

If this field is left blank, this policy is valid for all realms.

client

This is the requesting client, for which this action is valid. I.e. you can define different policies if the user access is allowed to manage his tokens from different IP addresses like the internal network or remotely via the firewall.

You can enter several IP addresses or subnets divided by comma (like 10.2.0.0/16, 192.168.0.1).

time

(added in privacyIDEA 2.12)

In the time field of a policy you can define a list of time ranges. A time range can consist of day of weeks (dow) and of times in 24h format. Possible values are:

<dow>: <hh>-<hh> <dow>: <hh:mm>-<hh:mm> <dow>-<dow>: <hh:mm>-<hh:mm>

You may use any combination of these. Like:

Mon-Fri: 8-18

to define certain policies to be active throughout working hours.

Note

If the time of a policy does not match, the policy is not found. Thus you can get effects you did not plan. So think at least twice before using time restricted policies.

Event Handler

Added in version 2.12.

What is the difference between Policies and event handlers?

Policies are used to define the behaviour of the system. With policies you can change the way the system reacts.

With event handlers you do not change the way the system reacts. But on certain events you can trigger a new action in addition to the behaviour defined in the policies.

These additional actions are also logged to the audit log. These actions are marked as EVENT in the audit log and you can see, which event triggered these actions. Thus a single API call can cause several audit log entries: One for the API call and more for the triggered actions.

Events

Each API call is an event and you can bind arbitrary actions to each event as you like.

Internally events are marked by a decorator “event” with an event identifier. At the moment not all events might be tagged. Please drop us a note to tag all further API calls.

An action is bound to the event token_init.

Handler Modules and Actions

The actions are defined in handler modules. So you bind a handler module and the action, defined in the handler module, to the events.

The handler module can define several actions and each action in the handler module can require additional options.

The event sendmail requires the option emailconfig.

Conditions

Added in version 2.14

An event handler module may also contain conditions. Only if all conditions are fullfilled, the action is triggered. Conditions are defined in the class property conditions and checked in the method check_condition. The UserNotification Event Handler defines such conditions.

Basic conditions

The basic event handler module has the following conditions.

last_auth

This condition checks if the last authentication is older than the specified time delta. The timedelta is specified with “h” (hours), “d” (days) or “y” (years). Specifying 180d would mean, that the action is triggered if the last successful authentication witht he token was berformed more than 180 days ago.

This can be used to send notifications to users or administrators to inform them, that there is a token, that might be orphaned.

logged_in_user

This condition checks if the logged in user is either an administrator or a normal user. This way the administrator can bind actions to events triggered by normal users or e.g. by help desk users. If a help desk user enrolls a token for a user, the user might get notified.

If a normal user enrolls some kind of token, the administrator might get notified.

otp_counter

The action is triggered, if the otp counter of a token has reached the given value.

The administrator can use this condition to e.g. automatically enroll a new paper token for the user or notify the user that nearly all OTP values of a paper token have been spent.

realm

The condition realm matches the user realm. The action will only trigger, if the user in this event is located in the given realm.

This way the administrator can bind certain actions to specific realms. E.g. some actions will only be triggered, if the event happens for normal users, but not for users in admin- or helpdesk realms.

result_value

This condition checks the result of an event.

E.g. the result of the event validate_check can be a failed authentication. This can be the trigger to notify either the token owner or the administrator.

serial

The action will only be triggered, if the serial number of the token in the event does match the regular expression.

This is a good idea to combine with other conditions. E.g. only tokens with a certain kind of serial number like Google Authenticator will be deleted automatically.

tokenrealm

In contrast to the realm this is the realm of the token - the tokenrealm. The action is only triggerd, if the token within the event has the given tokenrealm. This can be used in workflows, when e.g. hardware tokens which are not assigned to a user are pushed into a kind of storage realm.

tokentype

The action is only triggered if the token in this event is of the given type. This way the administrator can design workflows for enrolling and reenrolling tokens. E.g. the tokentype can be a registration token and the registration code can be easily and automatically sent to the user.

token_locked

The action is only triggered, if the token in the event is locked, i.e. the maximum failcounter is reached. In such a case the user can not use the token to authenticate anymore. So an action to notify the user or enroll a new token can be triggered.

token_has_owner

The action is only triggered, if the token is or is not assigned to a user.

token_is_orphaned

The action is only triggered, if the user, to whom the token is assigned, does not exist anymore.

This can be used to trigger the deletion of the token, if the token owner was removed from the userstore.

token_validity_period

Checks if the token is in the current validity period or not. Can be set to True or False.

Note

token_validity_period==False will trigger an action if either the validitiy period is either over or has not started, yet.

user_token_number

The action is only triggered, if the user in the event has the given number of tokens assigned.

This can be used to e.g. automatically enroll a token for the user if the user has no tokens left (token_number == 0) of to notify the administrator if the user has to many tokens assigned.

tokeninfo

The tokeninfo condition can compare any arbitrary tokeninfo field against a fixed value. You can compare strings and integers. Integers are converted automatically. Valid compares are:

myValue == 1000 myValue > 1000 myValue < 99 myTokenInfoField == EnrollmentState myTokenInfoField < ABC myTokenInfoField > abc

“myValue” and “myTokenInfoField” being any possible tokeninfo fields.

Starting with version 2.20 you can also compare dates in the isoformat like that:

myValue > 2017-10-12T10:00+0200 myValue < 2020-01-01T00:00+0000

In addition you can also use the tag {now} to compare to the curren time and you can add offsets to {now} in seconds, minutes, hours or days:

myValue < {now} myValue > {now}+10d myValue < {now}-5h

Which would match if the tokeninfo myValue is a date, which is later than 10 days from now or it the tokeninfo myValue is a date, which is 5 more than 5 hours in the past.

detail_error_message

This condition checks a regular expression against the detail section in the HTTP response. The field detail->error->message is evaluated.

Error messages can be manyfold. In case of authentication you could get error messages like:

“The user can not be found in any resolver in this realm!”

With token/init you could get:

“missing Authorization header”

..note:: The field ``detail->error->message is only available in case of an

internal error, i.e. if the response status is False.

detail_message

This condition checks a regular expression against the detail section in the HTTP response. The field detail->message is evaluated.

Those messages can be manyfold like:

“wrong otp pin”

“wrong otp value”

“Only 2 failed authentications per 1:00:00”

..note:: The field detail->message is available in case of status True,

like an authentication request that was handled successfully but failed.

Available Handler Modules

User Notification Handler Module

The user notification handler module is used to send emails token owners or administrators in case of any event.

Possible Actions

sendmail

The sendmail action sends an email to the tokenowner user. The email is sent, if an administrator managed the users token.

emailconfig

• required Option
• The email is sent via this SMTP server configuration.

subject

• optional

The subject line of the mail that is sent.

sendsms

The sendsms action sends an SMS to the tokenowner. The SMS is sent, if an administrator managed the users token.

smsconfig

• required Option
• The SMS Gateway configuration.

Options for both actions

Both actions sendmail and sendsms take several common options.

body

• optional

Here the administartor can specify the body of the email, that is sent. The body may contain the following tags

• {admin} name of the logged in user.
• {realm} realm of the logged in user.
• {action} the action that the logged in user performed.
• {serial} the serial number of the token.
• {url} the URL of the privacyIDEA system.
• {user} the given name of the token owner.
• {givenname} the given name of the token owner.
• {surname} the surname of the token owner.
• {username} the loginname of the token owner.
• {userrealm} the realm of the token owner.
• {tokentype} the type of the token.
• {registrationcode} the registration code in the detail response.
• {recipient_givenname} the given name of the recipient.
• {recipient_surname} the surname of the recipient.
• {googleurl_value} is the KEY URI for a google authenticator.
• {googleurl_img} is the data image source of the google authenticator QR code.
• {time} the current server time in the format HH:MM:SS.
• {date} the current server date in the format YYYY-MM-DD
• {client_ip} the client IP of the client, which issued the original request.
• {ua_browser} the user agent of the client, which issued the original request.
• {ua_string} the complete user agent string (including version number), which issued the original request.

mimetype

You can choose if the email should be sent as plain text or HTML. If the email is sent as HTML, you can do the following:

<a href={googleurl_value}>Your new token</a>

Which will create a clickable link. Clicked on the smartphone, the token will be imported to the smartphone app.

You can also do this:

<img src={googleurl_img}>

This will add the QR Code into the HTML email.

Warning

The KEY URI and the QR Code contain the secret OTP key in plain text. Everyone who receives this data has a detailed copy of this token. Thus we very much recommend to never send these data in an unencrypted email!

To

• required

This specifies to which type of user the notification should be sent. Possible recipient types are:

• token owner,
• logged in user,
• admin realm,
• internal admin,
• email address.

Depending on the recipient type you can enter additional information. The recipient type email takes a comma separated list of email addresses.

Code

This is the event handler module for user notifications. It can be bound to each event and can perform the action:

• sendmail: Send an email to the user/token owner
• sendsms: We can also notify the user with an SMS.

The module is tested in tests/test_lib_events.py

class privacyidea.lib.eventhandler.usernotification.NOTIFY_TYPE

Allowed token owner

ADMIN_REALM = 'admin realm'

EMAIL = 'email'

INTERNAL_ADMIN = 'internal admin'

LOGGED_IN_USER = 'logged_in_user'

TOKENOWNER = 'tokenowner'

class privacyidea.lib.eventhandler.usernotification.UserNotificationEventHandler

An Eventhandler needs to return a list of actions, which it can handle.

It also returns a list of allowed action and conditions

It returns an identifier, which can be used in the eventhandlig definitions

actions

This method returns a dictionary of allowed actions and possible options in this handler module.

Returns:dict with actions

description = 'This eventhandler notifies the user about actions on his tokens'

do(action, options=None)

This method executes the defined action in the given event.

Parameters:

• action –
• options (dict) – Contains the flask parameters g, request, response and the handler_def configuration

Returns:

identifier = 'UserNotification'

Token Handler Module

The token event handler module is used to perform actions on tokens in certain events.

This way you can define workflows to automatically modify tokens, delete or even create new tokens.

Possible Actions

set tokenrealm

Here you can set the token realms of the token.

E.g. You could use this action to automatically put all newly enrolled tokens

into a special realm by attaching this action to the event token_init.

delete

The token which was identified in the request will be deleted if all conditions are matched.

unassign

The token which was identified in the request will be unassign from the user if all conditions are matched.

disable

The token which was identified in the request will be disabled if all conditions are matched.

enable

The token which was identified in the request will be enabled if all conditions are matched.

enroll

If all conditions are matched a new token will be enrolled. This new token can be assigned to a user, which was identified in the request.

The administrator can specify the tokentype and the realms of the new

token.

set description

If all conditions are matched the description of the token identified in the request will be set.

You can use the tag {current_time} or {now} to set the current timestamp. In addition you can append an offset to current_time or now like {now}-12d or {now}+10m. This would write a timestamp which is 12 days in the passt or 10 minutes in the future. The plus or minus must follow without blank, allowed time identifiers are s (seconds), m (minutes), h (hours) and d (days).

Other tags are {client_ip} for the client IP address and {ua_browser} and {ua_string} for information on the user agent.

set validity

If all conditions are matched the validity period of the token will be set.

There are different possibilities to set the start and the end of the validity period. The event definition can either contain a fixed date and time or if can contain a time offset.

Fixed Time

A fixed time can be specified in the following formats.

Only date without time:

• 2016/12/23
• 23.12.2016

Date with time:

• 2016/12/23 9:30am
• 2016/12/23 11:20:pm
• 23.12.2016 9:30
• 23.12.2016 23:20

Starting with version 2.19 we recommend setting the fixed time in the ISO 8601 corresponding time format

• 2016-12-23T15:30+0600

Time Offset

You can also specify a time offset. In this case the validity period will be set such many days after the event occurred. This is indicated by using a “+” and a specifier for days (d), hours (h) and minutes (m).

E.g. +30m will set to start the validity period in 30 minutes after the event occurred.

+30d could set the validity period to end 30 days after an event occurred.

Note

This way you could easily define a event definition, which will set newly enrolled tokens to be only valid for a certain amount of days.

set countwindow

Here the count window of a token can be set. This requires an integer value.

set tokeninfo

Using the action set tokeninfo you can set any arbitrary tokeninfo attribute for the token. You need to specify the key of the tokeninfo and the value.

In the value field you can use the tag {current_time} to set the current timestamp. In addition you can append an offset to current_time or now like {now}-12d or {now}+10m. This would write a timestamp which is 12 days in the passt or 10 minutes in the future. The plus or minus must follow without blank, allowed time identifiers are s (seconds), m (minutes), h (hours) and d (days).

Other tags are {client_ip} for the client IP address and {ua_browser} and {ua_string} for information on the user agent and {username} and {realm} for information on the user in the parameters.

Note

Some tokens have token specific attributes that are stored in the tokeninfo. The TOTP token type has a timeWindow. The TOTP and the HOTP token store the hashlib in the tokeninfo, the SMS token stores the phone number.

Note

You can use this to set the timeWindow of a TOTP token for Automatic initial synchronization.

set failcounter

Using the action set failcounter you can reset the fail counter by setting it to 0 or also “block” the token by setting the fail counter to what

ever value the “max_fail” is, e.g. 10. Only integer values are allowed.

Code

This is the event handler module for token actions. You can attach token actions like enable, disable, delete, unassign,... of the

• current token
• all the user’s tokens
• all unassigned tokens
• all disabled tokens
• ...

class privacyidea.lib.eventhandler.tokenhandler.ACTION_TYPE

Allowed actions

DELETE = 'delete'

DISABLE = 'disable'

ENABLE = 'enable'

INIT = 'enroll'

SET_COUNTWINDOW = 'set countwindow'

SET_DESCRIPTION = 'set description'

SET_FAILCOUNTER = 'set failcounter'

SET_TOKENINFO = 'set tokeninfo'

SET_TOKENREALM = 'set tokenrealm'

SET_VALIDITY = 'set validity'

UNASSIGN = 'unassign'

class privacyidea.lib.eventhandler.tokenhandler.TokenEventHandler

An Eventhandler needs to return a list of actions, which it can handle.

It also returns a list of allowed action and conditions

It returns an identifier, which can be used in the eventhandlig definitions

actions

This method returns a dictionary of allowed actions and possible options in this handler module.

Returns:dict with actions

description = 'This event handler can trigger new actions on tokens.'

do(action, options=None)

This method executes the defined action in the given event.

Parameters:

• action –
• options (dict) – Contains the flask parameters g, request, response and the handler_def configuration

Returns:

identifier = 'Token'

class privacyidea.lib.eventhandler.tokenhandler.VALIDITY

Allowed validity options

END = 'valid till'

START = 'valid from'

Script Handler Module

The script event handler module is used to trigger external scripts in case of certain events.

This way you can even add external actions to your workflows. You could trigger a database dump, an external printing device, a backup and much more.

Possible Actions

The actions of the script event handler are the scripts located in a certain script directory. The default script directory is /etc/privacyidea/scripts.

You can change the location of the script directory and give the new directory in the parameter PI_SCRIPT_HANDLER_DIRECTORY in your pi.cfg

file.

Possible Options

Options can be passed to the script. Your script has to take care of the parsing of these parameters.

logged_in_role

Add the role of the logged in user. This can be either admin or user. If there is no logged in user, none will be passed.

The script will be called with the parameter

–logged_in_role <role>

logged_in_user

Add the logged in user. If there is no logged in user, none will be passed.

The script will be called with the parameter

–logged_in_user <username>@<realm>

realm

Add --realm <realm> as script parameter. If no realm is given, none will be passed.

serial

Add --serial <serial number> as script parameter. If no serial number is given, none will be passed.

user

Add --serial <username>' as script parameter. If no username is given, none will be passed.

Note

A possible script you could call is the privacyidea-get-unused-tokens.

Counter Handler Module

The counter event handler module is used to count certain events. You can define arbitrary counter names and each occurrence of an event will increase the counter in the counter table.

These counters can be used to graph time series of failed authentication, assigned tokens, user numbers or any other data with any condition over time.

Possible Actions

increase_counter

This action increases the counter in the database table eventcounter.

Possible Options

counter_name

This is the name of the counter in the database. You can have as many counters in as many event handlers as you like.

Federation Handler Module

The federation event handler can be used to configure relations between several privacyIDEA instances. Requests can be forwarded to child privacyIDEA instances.

Note

The federation event handler can modify the original response. If the response was modified a new field origin will be added to the detail section in the response. The origin will contain the URL of the privacyIDEA server that finally handled the request.

Possible Actions

forward

A request (usually an authentication request validate_check) can be forwarded to another privacyIDEA instance. The administrator can define privacyIDEA instances centrally at conifg -> privacyIDEA servers.

In addition to the privacyIDEA instance the action forward takes the following parameters:

client_ip The originial client IP will be passed to the child privacyIDEA server. Otherwise the child privacyIDEA server will use the parent privacyIDEA server as client.

Note

You need to configure the allow override client in the child privacyIDEA server.

realm The forwarding request will change the realm to the specified realm . This might be necessary since the child privacyIDEA server could have

different realms than the parent privacyIDEA server.

resolver The forwarding request will change the resolver to the specified

resolver. This might be necessary since the child privacyIDEA server could have different resolvers than the parent privacyIDEA server.

One simple possibility would be, that a user has a token in the parent privacyIDEA server and in the child privacyIDEA server. Configuring a forward event handler on the parent with the condition result_value = False would have the effect, that the user can either authenticate with the parent’s token or with the child’s token on the parent privacyIDEA server.

Federation can be used, if privacyIDEA was introduced in a subdivision of a larger company. When privacyIDEA should be enrolled to the complete company you can use federation. Instead of dropping the privacyIDEA instance in the subdivision and installing on single central privacyIDEA, the subdivision can still go on using the original privacyIDEA system (child) and the company will install a new top level privacyIDEA system (parent).

Using the federation handler you can setup many other, different scenarios we can not think of, yet.

Code

This is the event handler module for privacyIDEA federations. Requests can be forwarded to other privacyIDEA servers.

class privacyidea.lib.eventhandler.federationhandler.ACTION_TYPE

Allowed actions

FORWARD = 'forward'

class privacyidea.lib.eventhandler.federationhandler.FederationEventHandler

An Eventhandler needs to return a list of actions, which it can handle.

It also returns a list of allowed action and conditions

It returns an identifier, which can be used in the eventhandlig definitions

actions

This method returns a dictionary of allowed actions and possible options in this handler module.

Returns:dict with actions

description = 'This event handler can forward the request to other privacyIDEA servers'

do(action, options=None)

This method executes the defined action in the given event.

Parameters:

• action –
• options (dict) – Contains the flask parameters g, request, response and the handler_def configuration

Returns:

identifier = 'Federation'

Audit

The systems provides a sophisticated audit log, that can be viewed in the WebUI.

Audit Log

privacyIDEA comes with an SQL audit module. (see Audit log)

Cleaning up entries

The sqlaudit module writes audit entries to an SQL database. For performance reasons the audit module does no log rotation during the logging process.

But you can set up a cron job to clean up old audit entries. Since version 2.19 audit entries can be either cleaned up based on the number of entries or based on on the age.

Cleaning based on the age takes precedence:

You can specify a highwatermark and a lowwatermark. To clean up the audit log table, you can call pi-manage at command line:

pi-manage rotate_audit --highwatermark 20000 --lowwatermark 18000

This will, if there are more than 20.000 log entries, clean all old log entries, so that only 18000 log entries remain.

Cleaning based on the age:

You can specify the number of days, how old an audit entry may be at a max.

pi-manage rotate_audit –age 365

will delete all audit entries that are older than one year.

Cleaning based on the config file:

Using a config file you can define different retention times for the audit data. E.g. this way you can define, that audit entries about token listings can be deleted after one month, while the audit information about token creation will only deleted after ten years.

The config file is a YAML format and looks like this:

# DELETE auth requests of nils after 10 days
- rotate: 10
  user: nils
  action: .*/validate/check.*

# DELETE auth requests of friedrich after 7 days
- rotate: 7
  user: friedrich
  action: .*/validate/check.*

# Delete nagios user test auth directly
- rotate: 0
  user: nagiosuser
  action: POST /validate/check.*

# Delete token listing after one month
- rotate: 30
  action: ^GET /token

# Delete audit logs for token creating after 10 years
- rotate: 3650
  action: POST /token/init

# Delete everything else after 6 months
- rotate: 180
  action: .*

This is a list of rules. privacyIDEA iterates over all audit entries. The first matching rule for an entry wins. If the rule matches, the audit entry is deleted if the entry is older than the days specified in “rotate”.

If is a good idea to have a catch-all rule at the end.

Note

The keys “user”, “action”... correspond to the column names of the audit table. You can use any column name here like “date”, “action”, “action_detail”, “success”, “serial”, “administrator”, “user”, “realm”... for a complete list see the model definition. You may use Python regular expressions for matching.

You can the add a call like

pi-manage rotate_audit –config /etc/privacyidea/audit.yaml

in your crontab.

Access rights

You may also want to run the cron job with reduced rights. I.e. a user who has no read access to the original pi.cfg file, since this job does not need read access to the SECRET or PEPPER in the pi.cfg file.

So you can simply specify a config file with only the content:

PI_AUDIT_SQL_URI = <your database uri>

Then you can call pi-manage like this:

PRIVACYIDEA_CONFIGFILE=/home/cornelius/src/privacyidea/audit.cfg \
pi-manage rotate_audit

This will read the configuration (only the database uri) from the config file audit.cfg.

Table size

Sometimes the entires to be written to the database may be longer than the column in the database. You can either enlarge the columns in the database or you can set

PI_AUDIT_SQL_TRUNCATE = True

in pi.cfg. This will truncate each entry to the defined column length.

Client machines

privacyIDEA lets you define Machine Resolvers to connect to existing machine stores. The idea is for users to be able to authenticate on those client machines. Not in all cases an online authentication request is possible, so that authentication items can be passed to those client machines.

In addition you need to define, which application on the client machine the user should authenticate to. Different application require different authentication items.

Therefore privacyIDEA can define application types. At the moment privacyIDEA knows the application luks, offline and ssh. You can write your own application class, which is defined in Application Class.

You need to assign an application and a token to a client machine. Each application type can work with certain token types and each application type can use additional parameters.

Note

Not all tokens work well with all applications!

SSH

Currently working token types: SSH

Parameters:

user (optional, default=root)

When the SSH token type is assigned to a client, the user specified in the user parameter can login with the private key of the SSH token.

In the sshd_config file you need to configure the AuthorizedKeysCommand. Set it to:

privacyidea-authorizedkeys

This will fetch the SSH public keys for the requesting machine.

The command expects a configuration file /etc/privacyidea/authorizedkeyscommand which looks like this:

[Default]
url=https://localhost
admin=admin
password=test
nosslcheck=False

Note

To disable a SSH key for all servers, you simple can disable the SSH token in privacyIDEA.

Warning

In a productive environment you should not set nosslcheck to true, otherwise you are vulnerable to man in the middle attacks.

LUKS

Currently working token types: Yubikey Challenge Response

Parameters:

slot The slot to which the authentication information should be written

partition The encrypted partition (usually /dev/sda3 or /dev/sda5)

These authentication items need to be pulled on the client machine from the privacyIDEA server.

Thus, the following script need to be executed with root rights (able to write to LUKS) on the client machine:

privacyidea-luks-assign @secrets.txt --clearslot --name salt-minion

For more information please see the man page of this tool.

Offline

Currently working token types: HOTP.

Parameters:

user The local user, who should authenticate. (Only needed when calling machine/get_auth_items)

count The number of OTP values passed to the client.

The offline application also triggers when the client calls a /validate/check. If the user authenticates successfully with the correct token (serial number) and this very token is attached to the machine with an offline application the response to validate/check is enriched with a “auth_items” tree containing the salted SHA512 hashes of the next OTP values.

The client can cache these values to enable offline authentication. The caching is implemented in the privacyIDEA PAM module.

The server increases the counter to the last offline cached OTP value, so that it will not be possible to authenticate with those OTP values available offline on the client side.

Workflows and Tools

Import

Seed files that contain the secret keys of hardware tokens can be imported to the system via the menu Import.

The default import options are to import SafeNet XML file, OATH CSV files, Yubikey CSV files or PSKC files.

GPG Encryption

Starting with privacyIDEA 2.14 you can import GPG encrypted seed files. All files mentioned below can be encrypted this way.

privacyIDEA needs its own GPG key. You may create one like this:

mkdir /etc/privacyidea/gpg
GNUPGHOME=/etc/privacyidea/gpg gpg --gen-key

Then make sure, that the directory /etc/privacyidea/gpg is chown 700 for the user privacyidea.

Now you can export the public key and hand it to your token vendor:

GNUPGHOME=/etc/privacyidea/gpg gpg -a --export <keyid>

Now the token vendor can send the seed file GPG encrypted. You do not need to decrypt the file and store the decrypted file on a network folder. Just import the GPG encrypted file to privacyIDEA!

Note

Using the key PI_GNUPG_HOME in pi.cfg you can change the default above mentioned GNUPGHOME directory.

Note

privacyIDEA imports an ASCII armored file. The file needs to be encrypted like this:

gpg -e -a -r <keyid> import.csv

OATH CSV

This is a very simple CSV file to import HOTP, TOTP or OATH tokens. You can also convert your seed easily to this file format, to import the tokens.

The file format looks like this:

<serial>, <seed>, <type>, <otp length>, <time step>

For OCRA tokens it looks like this:

<serial>, <seed>, OCRA, <ocra suite>

serial is the serial number of the token that will also be used to identify the token in the database. Importing the same serial number twice will overwrite the token data.

seed is the secret key, that is used to calculate the OTP value. The seed is provided in a hexadecimal notation. Depending on the length either the SHA1 or SHA256 hash algorithm is identified.

type is either HOTP, TOTP or OCRA.

otp length is the length of the OTP value generated by the token. This is usually 6 or 8.

time step is the time step of TOTP tokens. This is usually 30 or 60.

ocra suite is the ocra suite of the OCRA token according to [1].

Yubikey CSV

Here you can import the CSV file that is written by the Yubikey personalization tool [2]. privacyIDEA can import all Yubikey modes, either Yubico mode or HOTP mode.

Note

There is an annoying drawback of the personalization tool: If you a initializing several HOTP yubikeys it will not write the serial number to the file.

PSKC

The Portable Symmetric Key Container is specified in [3]. OATH compliant token vendors provide the token seeds in a PSKC file. privacyIDEA lets you import PSKC files. All necessary information (OTP length, Hash algorithm, token type) are read from the file.

[1]	http://tools.ietf.org/html/rfc6287#section-6

[2]	http://www.yubico.com/products/services-software/personalization-tools/use/

[3]	https://tools.ietf.org/html/rfc6030

Token Enrollment Wizard

The enrollment wizard helps the user to enroll his first token. When enrolling the first token, we assume, that the user is not very familiar with the privacyIDEA web UI. So the enrollment wizard only contains a very reduced API.

Necessary requirements for the enrollment wizard

• The enrollment wizard will only be displayed, if the user has no token assigned, yet. Thus the user must be able to login to the web UI with his userstore password. This is the default behaviour or set the corresponding policy.
• Set a policy in scope webui and activate the policy action tokenwizard.
• The user will not be able to choose a token type. But the default token type will be enrolled.

You can see the token enrollment wizard in action here: https://www.youtube.com/watch?v=diAGbsiG8_A

Customization

There are two dialog windows in the wizard. You can configure the text in the wizard in your html templates defined in these files:

static/customize/views/includes/token.enroll.pre.top.html static/customize/views/includes/token.enroll.pre.bottom.html static/customize/views/includes/token.enroll.post.top.html static/customize/views/includes/token.enroll.post.bottom.html

Note

You can change the directory static/customize to a URL that fits your needs the best by defining a variable PI_CUSTOMIZATION in the file pi.cfg. This way you can put all modifications in one place apart from the original code.

Example

Your privacyIDEA system is running in the URL sub path /pi. The files could be addressed via a path component mydesign (in this case pi/mydesign). Thus the WebUI will look for the files in the URL path /pi/mydesign/views/includes/.

So you set in pi.cfg:

PI_CUSTOMIZATION = “/mydesign”

Your customized files are located in /etc/privacyidea/customize/views/includes/. In the Apache webserver you need to map /pi/mydesign to /etc/privacyidea/customize:

Alias /pi/mydesign /etc/privacyidea/customize

Tools

privacyIDEA comes with a list of command line tools, which also help to automate tasks.

privacyidea-token-janitor

Starting with version 2.19 privacyIDEA comes with a token janitor script. This script can find orphaned tokens, unused tokens or tokens of specific type, description or token info.

It can unassign, delete or disable those tokens and it can set additional tokeninfo or descriptions.

If you are unsure to directly delete orphaned tokens, because there might be a glimpse in the connection to your user store, you could as well in a first step mark the orphaned tokens. A day later you could run the script again and delete those tokens, which are (still) orphaned and marked.

privacyidea-get-unused-tokens

The script privacyidea-get-unused-tokens allows you to search for tokens, which were not used for authentication for a while. These tokens can be listed, disabled, marked or deleted.

You can specify how old the last authentication of such a token has to be. You can use the tags h (hours), d (day) and y (year). Sepcifying 180d will find tokens, that were not used for authentication for the last 180 days.

The command

privacyidea-get-unused-tokens disable 180d

will disable those tokens.

This script can be well used with the Script Handler Module.

Two Step Enrollment

Starting with version 2.21 privacyIDEA allows to enroll smartphone based tokens in a 2step enrollment.

With the rise of the smartphones and the fact that every user has a smartphone, carries it with him all the time and cares about it a lot, using the smartphone for authentication gets more and more attractive to IT departments.

Google came up with the Key URI [1] to use a QR code to easily enroll a smartphone token, i.e. transport the OTP secret from the server to the phone. However this bears some security issues as already pointed out [2].

This is why privacyIDEA allows to generate the OTP secret from a server component and from a client component (generated by the smartphone). This way the enrolled token is more tightly bound to this single smartphone and can not be copied that easily anymore.

Workflow

In a two step enrollment process the user clicks in the Web UI to enroll a token. The server generates a QR code and the user will scan this QR code with his smartphone app. The QR code contains the server component of the key and the information, that a second component is needed.

The smartphone generates the second component and displays this to the user.

The user enters this second component into the privacyIDEA Web UI.

Both the smartphone and the server calculate the OTP secret from both components.

Two Step policies

Two step enrollment is controlled by policies in the admin/user scope and in the enrollment scope.

Thus the administrator can allow or force a user (or other administrators) to do a two step enrollment. This way it is possible to avoid the enrollment of insecure Google Authenticator QR codes in the complete installation. (hotp_2step and totp_2step).

The default behaviour is to not allow a two step enrollment. Only if a corresponding admin or user policy is defined, two step enrollment is possible.

Key generation

In addition the administrator can define an enrollment policy to specify necessary parameters for the key generation.

Two step enrollment is possible for HOTP and TOTP tokens. Thus the administrator can define token type specific policies in the scope enrollment: hotp_2step_clientsize, totp_2step_clientsize, hotp_2step_difficulty... see {type}_2step_clientsize, {type}_2step_serversize, {type}_2step_difficulty.

privacyIDEA Authenticator

The privacyIDEA Authenticator [3] that is available from the Google Play Store supports the two step enrollment.

Specification

The two step enrollment simply adds some parameters to the original Key URI.

2step_output

This is the resulting key size, which the smartphone should generate (in bytes).

2step_salt

This is the length of the client component that the smartphone should generate (in bytes).

2step_difficulty

This is the number of rounds for the PBKDF2 that the smartphone should use to generate the OTP secret.

The secret parameter of the Key URI contains the server component.

The smartphone app then generates the client component, which is 2step_salt random bytes. It is then displayed in a human-readable format called base32check:

b32encode(sha1(client_component).digest()[0:4] + client_component).strip("=")

In other words, the first four bytes of the client component’s SHA-1 hash are concatenated with the actual client component. The result is encoded using base32, whereas trailing padding characters are removed.

The second step of the enrollment process is realized as another request to the /token/init endpoint:

POST /token/init

serial=<token serial>
otpkey=<base32check(client_component)>
otpkeyformat=base32check

Server and smartphone app then use PBKDF2 to generate the final secret (see [4] for parameter names):

secret = PBKDF2(P=hexlify(<server component>),
                S=<client component>,
                c=<2step_difficulty>
                dkLen=<2step_output>)

whereas hexlify(<server component>) denotes a hex-encoding (using lowercase letters) of the byte array which comprises the server component.

Note

Please note that the two-step enrollment process is currently not designed to protect against malicious attackers. Depending on the choice of iteration count and salt size, an attacker who knows the server component and an OTP value may be able to obtain the client component with a brute-force approach. However, two-step enrollment is still an improvement to the status quo, as a simple copy of the QR code does not immediately leak the OTP secret and obtaining the OTP secret using brute-force is not trivial.

[1]	https://github.com/google/google-authenticator/wiki/Key-Uri-Format

[2]	https://netknights.it/en/the-problem-with-the-google-authenticator/

[3]	https://play.google.com/store/apps/details?id=it.netknights.piauthenticator

[4]	https://www.ietf.org/rfc/rfc2898.txt

Application Plugins

privacyIDEA comes with application plugins. These are plugins for applications like PAM, OTRS, Apache2, FreeRADIUS, ownCloud or simpleSAMLphp which enable these application to authenticate users against privacyIDEA.

You may also write your own application plugin or connect your own application to privacyIDEA. This is quite simple using a REST API Validate endpoints.

Pluggable Authentication Module

The PAM module of privacyIDEA directly communicates with the privacyIDEA server via the API. The PAM module also supports offline authentication. In this case you need to configure an offline machine application. (See Offline)

You can install the PAM module with a ready made Debian package for Ubuntu or just use the source code file. It is a python module, that requires pam-python.

The configuration could look like this:

... pam_python.so /path/to/privacyidea_pam.py
url=https://localhost prompt=privacyIDEA_Authentication

The URL parameter defaults to https://localhost. You can also add the parameters realm= and debug.

If you want to disable certificate validation, which you should not do in a productive environment, you can use the parameter nosslverify.

A new parameter cacerts= lets you define a CA Cert-Bundle file, that contains the trusted certificate authorities in PEM format.

The default behaviour is to trigger an online authentication request. If the request was successful, the user is logged in. If the request was done with a token defined for offline authentication, then in addition all offline information is passed to the client and cached on the client so that the token can be used to authenticate without the privacyIDEA server available.

try_first_pass

Starting with version 2.8 privacyidea_pam supports try_first_pass. In this case the password that exists in the PAM stack will be sent to privacyIDEA. If this password is successfully validated, than the user is logged in without additional requests. If the password is not validated by privacyIDEA, the user is asked for an additional OTP value.

Note

This can be used in conjunction with the passthru policy. In this case users with no tokens will be able to login with only the password in the PAM stack.

Read more about how to use PAM to do OTP with OpenVPN.

Using pam_yubico

If you are using yubikey tokens you might also use pam_yubico. You can use Yubikey tokens for two more or less distinct applications. The first is using privacyideas PAM module as described above. In this case privacyidea handles the policies for user access and password validation. This works fine, when you only use privacyidea for token validation.

The second mode is using the standard PAM module for yubikeys from Yubico pam_yubico to handle the token validation. The upside ist that you can use the PAM module included with you distribution, but there are downsides as well.

• You can’t set a token PIN in privacyidea, because pam_yubico tries to use the token PIN entered by the user as a system password (which is likely to fail), i.e. the PIN will be stripped by pam_yubico and will not reach the privacyIDEA system.
• Setting the policy which tokens are valid for which users is done either in ~/.yubico/authorized_keys or in the file given by the authfile option in the PAM configuration. The api server will only validate the token, but not check any kind of policy.

You can work around the restrictions by using a clever combination of tokentype yubikey and yubico as follows:

• enroll a yubikey token with yubikey_mass_enroll --mode YUBICO.
• do not set a token password.
• do not assign the token to a user.
• please make a note of yubikey.prefix (12 characters starting with vv).

Now the token can be used with pam_yubico, but will not allow any user access in privacyidea. If you want to use the token with pam_yubico see the manual page for details. You’ll want something like the following in your PAM config:

auth required pam_yubico.so id=<apiid> key=<API key> \
     urllist=https://<privacyidea-server>/ttype/yubikey authfile=/etc/yubikeys/authorized_yubikeys

The file /etc/yubikeys/authorized_yubikeys contains a line for each user with the username and the allowed tokens delimited by ”:”, for example:

<username>:<serial number1>:<prefix1>:<prefix2>

... doc/configuration/tokenconfig, add yubikey.rst to describe how to configure Client ID/apiid and API key

Now create a second token representing the Yubikey, but this time use the Yubico Cloud mode. Go to Tokens -> Enroll Token and select Yubico Cloud mode. Enter the 12 characters prefix you noted above and assign this token to a user and possibly set a token PIN. It would be nice to have the the serial number of the UBCM token correspond to the UBAM token, but this is right now not possible with the WebUI.

In the WebUI, test the UBAM token without a Token PIN, test the UBCM token with the stored Token PIN, and check the token info afterwards. Check the yubikey token via /ttype/yubikey, for example with:

ykclient --debug --url https://<privacyidea>/ttype/yubikey --apikey "<API key>" "apiid" <otp>

There should be successful authentications (count_auth_success), but no failures.

FreeRADIUS

Starting with privacyIDEA 2.19, there are two ways to integrate FreeRADIUS:

• Using a Perl-based privacyIDEA plugin, which is available for FreeRADIUS 2.0.x and above. It supports advanced use cases (such as challenge-response authentication or attribute mapping). Read more about it at RADIUS plugin.
• Using the rlm_rest plugin provided by FreeRADIUS 3.0.x and above. However, this setup does not support challenge-response or attribute mapping. Read more about it at Configuration of rlm_rest.

With either setup, you can test the RADIUS setup using a command like this:

echo "User-Name=user, Password=password" | radclient -sx yourRadiusServer \
   auth topsecret

Note

Do not forget to configure the clients.conf accordingly.

Microsoft NPS server

You can also use the Microsoft Network Protection Server with privacyIDEA. A full featured integration guide can be found at the NetKnights webpage [5].

simpleSAMLphp Plugin

You can install the plugin for simpleSAMLphp on Ubuntu 14.04 LTS (see SimpleSAMLphp) or on any other distribution using the source files from [1].

Follow the simpleSAMLphp instructions to configure your authsources.php. A usual configuration will look like this:

'example-privacyidea' => array(
    'privacyidea:privacyidea',

    /*
     * The name of the privacyidea server and the protocol
     * A port can be added by a colon
     * Required.
     */
    'privacyideaserver' => 'https://your.server.com',

    /*
     * Check if the hostname matches the name in the certificate
     * Optional.
     */
    'sslverifyhost' => False,

    /*
     * Check if the certificate is valid, signed by a trusted CA
     * Optional.
     */
    'sslverifypeer' => False,

    /*
     * The realm where the user is located in.
     * Optional.
     */
    'realm' => '',

    /*
     * This is the translation from privacyIDEA attribute names to
     * SAML attribute names.
     */
     'attributemap' => array('username' => 'samlLoginName',
                             'surname' => 'surName',
                             'givenname' => 'givenName',
                             'email' => 'emailAddress',
                             'phone' => 'telePhone',
                             'mobile' => 'mobilePhone',
                             ),
),

TYPO3

You can install the privacyIDEA extension from the TYPO3 Extension Repository. The privacyIDEA extension is easily configured.

privacyIDEA Server URL

This is the URL of your privacyIDEA installation. You do not need to add the path validate/check. Thus the URL for a common installation would be https://yourServer/.

Check certificate

Whether the validity of the SSL certificate should be checked or not.

Warning

If the SSL certificate is not checked, the authentication request could be modified and the answer to the request can be modified, easily granting access to an attacker.

Enable privacyIDEA for backend users

If checked, a user trying to authenticate at the backend, will need to authenticate against privacyIDEA.

Enable privacyIDEA for frontend users

If checked, a user trying to authenticate at the frontend, will need to authenticate against privacyIDEA.

Pass to other authentication module

If the authentication at privacyIDEA fails, the credential the user entered will be verified against the next authentication module.

This can come in handy, if you are setting up the system and if you want to avoid locking yourself out.

Anyway, in a productive environment you probably want to uncheck this feature.

OTRS

There are two plugins for OTRS. For OTRS version 4.0 and higher use privacyIDEA-4_0.pm.

This perl module needs to be installed to the directory Kernel/System/Auth.

On Ubuntu 14.04 LTS you can also install the module using the PPA repository and installing:

apt-get install privacyidea-otrs

To activate the OTP authentication you need to add the following to Kernel/Config.pm:

$Self->{'AuthModule'} = 'Kernel::System::Auth::privacyIDEA';
$Self->{'AuthModule::privacyIDEA::URL'} = \
        "https://localhost/validate/check";
$Self->{'AuthModule::privacyIDEA::disableSSLCheck'} = "yes";

Note

As mentioned earlier you should only disable the checking of the SSL certificate if you are in a test environment. For productive use you should never disable the SSL certificate checking.

Note

This plugin requires, that you also add the path validate/check to the URL.

Apache2

The Apache plugin uses mod_wsgi and redis to provide a basic authentication on Apache2 side and validating the credentials against privacyIDEA.

On Ubuntu 14.04 LTS you can easily install the module from the PPA repository by issuing:

apt-get install privacyidea-apache-client

To activate the OTP authentication on a “Location” or “Directory” you need to configure Apache2 like this:

<Directory /var/www/html/secretdir>
     AuthType Basic
     AuthName "Protected Area"
     AuthBasicProvider wsgi
     WSGIAuthUserScript /usr/share/pyshared/privacyidea_apache.py
     Require valid-user
</Directory>

Note

Basic Authentication sends the base64 encoded password on each request. So the browser will send the same one time password with each reqeust. Thus the authentication module needs to cache the password as the successful authentication. Redis is used for caching the password.

Warning

As redis per default is accessible by every user on the machine, you need to use this plugin with caution! Every user on the machine can access the redis database to read the passwords of the users. The cached credentials are stored as pbkdf2+sha512 hash.

NGINX

The NGINX plugin uses the internal scripting language lua of the NGINX webserver and redis as caching backend to provide basic authentication against privacyIDEA.

On Ubuntu 14.04 LTS or Debian Jessi 8 you can easyly install the module by installing the following packages:

nginx-extras lua-nginx-redis lua-cjson redis-server

You can retrieve the nginx plugin here: [4]

To activate the OTP authentication on a “Location” you need to include the lua script that basically verifies the given credentials against the caching backend. New authentications will be sent to a different (internal) location via subrequest which points to the privacyIDEA authentication backend (via proxy_pass).

For the basic configuration you need to include the following lines to your location block

location / {

# additional plugin configuration goes here # access_by_lua_file ‘privacyidea.lua’;

} location /privacyidea-validate-check {

internal; proxy_pass https://privacyidea/validate/check;

}

You can customize the authentication plugin by setting some of the following variables in the secured location block:

# redis host:port
# set $privacyidea_redis_host "127.0.0.1";
set $privacyidea_redis_post 6379;

# how long are accepted authentication allowed to be cached
# if expired, the user has to reauthenticate
set $privacyidea_ttl 900;

# privacyIDEA realm. leave empty == default
set $privacyidea_realm 'somerealm'; # (optional)

# pointer to the internal validation proxy pass
set $privacyidea_uri "/privacyidea-validate-check";

# the http realm presented to the user
set $privacyidea_http_realm "Secure zone (use PIN + OTP)";

Note

Basic Authentication sends the base64 encoded password on each request. So the browser will send the same one time password with each reqeust. Thus the authentication module needs to cache the password as the successful authentication. Redis is used for caching the password similar to the Apache2 plugin.

Warning

As redis per default is accessible by every user on the machine, you need to use this plugin with caution! Every user on the machine can access the redis database to read the passwords of the users. The cached credentials are stored as SHA1_HMAC hash. If you prefer a stronger hashing method feel free to extend the given password_hash/verify functions using additional lua libraries (for example by using lua-resty-string).

ownCloud

The ownCloud plugin is a ownCloud user backend. The directory user_privacyidea needs to be copied to your owncloud apps directory.

Activating the ownCloud plugin

You can then activate the privacyIDEA ownCloud plugin by checking Use privacyIDEA to authenticate the users. All users now need to be known to privacyIDEA and need to authenticate using the second factor enrolled in privacyIDEA - be it an OTP token, Google Authenticator or SMS/Smartphone.

Checking Also allow users to authenticate with their normal passwords. lets the user choose if he wants to authenticate with the OTP token or with his original password from the original user backend.

Note

At the moment using a desktop client with a one time password is not supported.

ownCloud 9.1 and Nextcloud 10 come with a new two factor framework. The new privacyIDEA ownCloud App allows you to add a second factor, that is centrally managed by privacyIDEA to the ownCloud or Nextcloud installation.

The ownCloud privacyIDEA App is available here [7].

The App requires a subscription file to work for more than ten users. You can

get the subscription file at NetKnights [8].

Django

You can add two factor authentication with privacyIDEA to Django using this Django plugin. See django.

You can simple add PrivacyIDEA class to AUTHENTICATION_BACKENDS settings of Django.

OpenVPN

Read more about how to use OpenVPN with privacyidea at OTP with OpenVPN.

Windows

Credential Provider

The privacyIDEA Credential Provider adds two factor authentication to the Windows desktop or Terminal server. See http://privacyidea-credential-provider.readthedocs.io

Provider Class

There is a dot Net provider class, which you can use to integrate privacyIDEA authentication into other products and worflows. See https://github.com/sbidy/privacyIDEA_dotnetProvider

Further plugins

You can find further plugins for Dokuwiki, Wordpress, Contao and Django at [3].

[1]	https://github.com/privacyidea/simplesamlphp-module-privacyidea

[2]	https://github.com/privacyidea/privacyidea/tree/master/authmodules

[3]	https://github.com/cornelinux?tab=repositories

[4]	https://github.com/dhoffend/lua-nginx-privacyidea

[5]	https://netknights.it/en/nps-2012-for-two-factor-authentication-with-privacyidea/

[6]	https://github.com/jeweber/django-privacyidea-auth

[7]	https://apps.owncloud.com/content/show.php/privacyIDEA+ownCloud+App?content=174779

[8]	https://netknights.it/en/produkte/privacyidea-owncloud-app/

Code Documentation

The code roughly has three levels.

API level

The API level is used to access the system. For some calls you need to be authenticated as administrator, for some calls you can be authenticated as normal user. These are the token and the audit endpoint. For calls to the validate API you do not need to be authenticated at all.

At this level Authentication is performed. In the lower levels there is no authentication anymore.

The object g.logged_in_user is used to pass the authenticated user. The client gets a JSON Web Token to authenticate every request.

API functions are decorated with the decorators admin_required and user_required to define access rules.

REST API

This is the REST API for privacyidea. It lets you create the system configuration, which is denoted in the system endpoints.

Special system configuration is the configuration of

• the resolvers
• the realms
• the defaultrealm
• the policies.

Resolvers are dynamic links to existing user sources. You can find users in LDAP directories, SQL databases, flat files or SCIM services. A resolver translates a loginname to a user object in the user source and back again. It is also responsible for fetching all additional needed information from the user source.

Realms are collections of resolvers that can be managed by administrators and where policies can be applied.

Defaultrealm is a special endpoint to define the default realm. The default realm is used if no user realm is specified. If a user from realm1 tries to authenticate or is addressed, the notation user@realm1 is used. If the @realm1 is omitted, the user is searched in the default realm.

Policies are rules how privacyidea behaves and which user and administrator is allowed to do what.

Start to read about authentication to the API at Authentication endpoints.

Now you can take a look at the several REST endpoints.

This REST API is used to authenticate the users. A user needs to authenticate when he wants to use the API for administrative tasks like enrolling a token.

This API must not be confused with the validate API, which is used to check, if a OTP value is valid. See Validate endpoints.

Authentication of users and admins is tested in tests/test_api_roles.py

You need to authenticate for all administrative tasks. If you are not authenticated, the API returns a 401 response.

To authenticate you need to send a POST request to /auth containing username and password.

Audit endpoint

GET /audit/statistics

get the statistics values from the audit log

JSON Parameters:  

• days – The number of days to run the stats
• start – The start time to run the stats
• end – The end time to run the stats

If start or end is missing, the days are used.

The time is to be passed in the format

yyyy-MM-ddTHH:mmZ

Example request:

GET /audit/statistics HTTP/1.1
Host: example.com
Accept: application/json

Example response:

HTTP/1.1 200 OK
Content-Type: text/csv

 {
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": [
       {
          "serial_plot": "...image data...",
       }
     ]
   },
   "version": "privacyIDEA unknown"
 }

GET /audit/

return a paginated list of audit entries.

Params can be passed as key-value-pairs.

Httpparam timelimit:  A timelimit, that limits the recent audit entries. This param gets overwritten by a policy auditlog_age. Can be 1d, 1m, 1h.

Example request:

GET /audit?realm=realm1 HTTP/1.1
Host: example.com
Accept: application/json

Example response:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": [
       {
          "serial": "....",
          "missing_line": "..."
       }
     ]
   },
   "version": "privacyIDEA unknown"
 }

GET /audit/(csvfile)

Download the audit entry as CSV file.

Params can be passed as key-value-pairs.

Example request:

GET /audit/audit.csv?realm=realm1 HTTP/1.1
Host: example.com
Accept: text/csv

Example response:

HTTP/1.1 200 OK
Content-Type: text/csv

 {
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": [
       {
          "serial": "....",
          "missing_line": "..."
       }
     ]
   },
   "version": "privacyIDEA unknown"
 }

Authentication endpoints

This REST API is used to authenticate the users. A user needs to authenticate when he wants to use the API for administrative tasks like enrolling a token.

This API must not be confused with the validate API, which is used to check, if a OTP value is valid. See Validate endpoints.

Authentication of users and admins is tested in tests/test_api_roles.py

You need to authenticate for all administrative tasks. If you are not authenticated, the API returns a 401 response.

To authenticate you need to send a POST request to /auth containing username and password.

GET /auth/rights

This returns the rights of the logged in user.

Request Headers:  

• Authorization – The authorization token acquired by /auth request

POST /auth

This call verifies the credentials of the user and issues an authentication token, that is used for the later API calls. The authentication token has a validity, that is usually 1 hour.

JSON Parameters:  

• username – The username of the user who wants to authenticate to the API.
• password – The password/credentials of the user who wants to authenticate to the API.

Return:

A json response with an authentication token, that needs to be used in any further request.

Status Codes:

• 200 OK – in case of success
• 401 Unauthorized – if authentication fails

Example Authentication Request:

POST /auth HTTP/1.1
Host: example.com
Accept: application/json

username=admin
password=topsecret

Example Authentication Response:

HTTP/1.0 200 OK
Content-Length: 354
Content-Type: application/json

{
     "id": 1,
     "jsonrpc": "2.0",
     "result": {
         "status": true,
         "value": {
             "token": "eyJhbGciOiJIUz....jdpn9kIjuGRnGejmbFbM"
         }
     },
     "version": "privacyIDEA unknown"
}

Response for failed authentication:

HTTP/1.1 401 UNAUTHORIZED
Content-Type: application/json
Content-Length: 203

{
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "error": {
       "code": -401,
       "message": "missing Authorization header"
     },
     "status": false
   },
   "version": "privacyIDEA unknown",
   "config": {
     "logout_time": 30
   }
}

Example Request:

Requests to privacyidea then should use this security token in the Authorization field in the header.

GET /users/ HTTP/1.1
Host: example.com
Accept: application/json
Authorization: eyJhbGciOiJIUz....jdpn9kIjuGRnGejmbFbM

Validate endpoints

This module contains the REST API for doing authentication. The methods are tested in the file tests/test_api_validate.py

Authentication is either done by providing a username and a password or a serial number and a password.

Authentication workflow

Authentication workflow is like this:

In case of authenticating a user:

• privacyidea.lib.token.check_user_pass()
• privacyidea.lib.token.check_token_list()
• privacyidea.lib.tokenclass.TokenClass.authenticate()
• privacyidea.lib.tokenclass.TokenClass.check_pin()
• privacyidea.lib.tokenclass.TokenClass.check_otp()

In case if authenitcating a serial number:

• privacyidea.lib.token.check_serial_pass()
• privacyidea.lib.token.check_token_list()
• privacyidea.lib.tokenclass.TokenClass.authenticate()
• privacyidea.lib.tokenclass.TokenClass.check_pin()
• privacyidea.lib.tokenclass.TokenClass.check_otp()

GET /validate/triggerchallenge

An administrator can call this endpoint if he has the right of triggerchallenge (scope: admin). He can pass a user name and or a serial number. privacyIDEA will trigger challenges for all native challenges response tokens, possessed by this user or only for the given serial number.

The request needs to contain a valid PI-Authorization header.

Parameters:

• user – The loginname/username of the user, who tries to authenticate.
• realm – The realm of the user, who tries to authenticate. If the realm is omitted, the user is looked up in the default realm.
• serial – The serial number of the token.

Return:

a json result with a “result” of the number of matching challenge response tokens

Example response for a successful triggering of challenge:

{"jsonrpc": "2.0",
 "signature": "1939...146964",
 "detail": {"transaction_ids": ["03921966357577766962"],
            "messages": ["Enter the OTP from the SMS:"],
            "threadid": 140422378276608},
 "versionnumber": "unknown",
 "version": "privacyIDEA unknown",
 "result": {"status": true,
            "value": 1},
 "time": 1482223663.517212,
 "id": 1}

Example response for response, if the user has no challenge token:

{"detail": {"messages": [],
            "threadid": 140031212377856,
            "transaction_ids": []},
 "id": 1,
 "jsonrpc": "2.0",
 "result": {"status": true,
            "value": 0},
 "signature": "205530282...54508",
 "time": 1484303812.346576,
 "version": "privacyIDEA 2.17",
 "versionnumber": "2.17"}

Example response for a failed triggering of a challenge. In this case

the status will be false.

{"detail": null,
 "id": 1,
 "jsonrpc": "2.0",
 "result": {"error": {"code": 905,
                      "message": "ERR905: The user can not be
                      found in any resolver in this realm!"},
            "status": false},
 "signature": "14468...081555",
 "time": 1484303933.72481,
 "version": "privacyIDEA 2.17"}

POST /validate/triggerchallenge

An administrator can call this endpoint if he has the right of triggerchallenge (scope: admin). He can pass a user name and or a serial number. privacyIDEA will trigger challenges for all native challenges response tokens, possessed by this user or only for the given serial number.

The request needs to contain a valid PI-Authorization header.

Parameters:

• user – The loginname/username of the user, who tries to authenticate.
• realm – The realm of the user, who tries to authenticate. If the realm is omitted, the user is looked up in the default realm.
• serial – The serial number of the token.

Return:

a json result with a “result” of the number of matching challenge response tokens

Example response for a successful triggering of challenge:

{"jsonrpc": "2.0",
 "signature": "1939...146964",
 "detail": {"transaction_ids": ["03921966357577766962"],
            "messages": ["Enter the OTP from the SMS:"],
            "threadid": 140422378276608},
 "versionnumber": "unknown",
 "version": "privacyIDEA unknown",
 "result": {"status": true,
            "value": 1},
 "time": 1482223663.517212,
 "id": 1}

Example response for response, if the user has no challenge token:

{"detail": {"messages": [],
            "threadid": 140031212377856,
            "transaction_ids": []},
 "id": 1,
 "jsonrpc": "2.0",
 "result": {"status": true,
            "value": 0},
 "signature": "205530282...54508",
 "time": 1484303812.346576,
 "version": "privacyIDEA 2.17",
 "versionnumber": "2.17"}

Example response for a failed triggering of a challenge. In this case

the status will be false.

{"detail": null,
 "id": 1,
 "jsonrpc": "2.0",
 "result": {"error": {"code": 905,
                      "message": "ERR905: The user can not be
                      found in any resolver in this realm!"},
            "status": false},
 "signature": "14468...081555",
 "time": 1484303933.72481,
 "version": "privacyIDEA 2.17"}

POST /validate/offlinerefill

This endpoint allows to fetch new offline OTP values for a token, that is already offline. According to the definition it will send the missing OTP values, so that the client will have as much otp values as defined.

Parameters:

• serial – The serial number of the token, that should be refilled.
• refilltoken – The authorization token, that allows refilling.
• pass – the last password (maybe password+OTP) entered by the user

Return:

GET /validate/radiuscheck

check the authentication for a user or a serial number. Either a serial or a user is required to authenticate. The PIN and OTP value is sent in the parameter pass. In case of successful authentication it returns result->value: true.

In case /validate/radiuscheck is requested, the responses are modified as follows: A successful authentication returns an empty HTTP 204 response. An unsuccessful authentication returns an empty HTTP 400 response. Error responses are the same responses as for the /validate/check endpoint.

Parameters:

• serial – The serial number of the token, that tries to authenticate.
• user – The loginname/username of the user, who tries to authenticate.
• realm – The realm of the user, who tries to authenticate. If the realm is omitted, the user is looked up in the default realm.
• pass – The password, that consists of the OTP PIN and the OTP value.
• otponly – If set to 1, only the OTP value is verified. This is used in the management UI. Only used with the parameter serial.
• transaction_id – The transaction ID for a response to a challenge request
• state – The state ID for a response to a challenge request

Return:

a json result with a boolean “result”: true

Example Validation Request:

POST /validate/check HTTP/1.1
Host: example.com
Accept: application/json

user=user
realm=realm1
pass=s3cret123456

Example response for a successful authentication:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "detail": {
     "message": "matching 1 tokens",
     "serial": "PISP0000AB00",
     "type": "spass"
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": true
   },
   "version": "privacyIDEA unknown"
 }

Example response for this first part of a challenge response authentication:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "detail": {
     "serial": "PIEM0000AB00",
     "type": "email",
     "transaction_id": "12345678901234567890",
     "multi_challenge: [ {"serial": "PIEM0000AB00",
                          "transaction_id":  "12345678901234567890",
                          "message": "Please enter otp from your
                          email"},
                         {"serial": "PISM12345678",
                          "transaction_id": "12345678901234567890",
                          "message": "Please enter otp from your
                          SMS"}
     ]
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": false
   },
   "version": "privacyIDEA unknown"
 }

In this example two challenges are triggered, one with an email and one with an SMS. The application and thus the user has to decide, which one to use. They can use either.

Note

All challenge response tokens have the same transaction_id in this case.

POST /validate/radiuscheck

check the authentication for a user or a serial number. Either a serial or a user is required to authenticate. The PIN and OTP value is sent in the parameter pass. In case of successful authentication it returns result->value: true.

In case /validate/radiuscheck is requested, the responses are modified as follows: A successful authentication returns an empty HTTP 204 response. An unsuccessful authentication returns an empty HTTP 400 response. Error responses are the same responses as for the /validate/check endpoint.

Parameters:

• serial – The serial number of the token, that tries to authenticate.
• user – The loginname/username of the user, who tries to authenticate.
• realm – The realm of the user, who tries to authenticate. If the realm is omitted, the user is looked up in the default realm.
• pass – The password, that consists of the OTP PIN and the OTP value.
• otponly – If set to 1, only the OTP value is verified. This is used in the management UI. Only used with the parameter serial.
• transaction_id – The transaction ID for a response to a challenge request
• state – The state ID for a response to a challenge request

Return:

a json result with a boolean “result”: true

Example Validation Request:

POST /validate/check HTTP/1.1
Host: example.com
Accept: application/json

user=user
realm=realm1
pass=s3cret123456

Example response for a successful authentication:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "detail": {
     "message": "matching 1 tokens",
     "serial": "PISP0000AB00",
     "type": "spass"
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": true
   },
   "version": "privacyIDEA unknown"
 }

Example response for this first part of a challenge response authentication:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "detail": {
     "serial": "PIEM0000AB00",
     "type": "email",
     "transaction_id": "12345678901234567890",
     "multi_challenge: [ {"serial": "PIEM0000AB00",
                          "transaction_id":  "12345678901234567890",
                          "message": "Please enter otp from your
                          email"},
                         {"serial": "PISM12345678",
                          "transaction_id": "12345678901234567890",
                          "message": "Please enter otp from your
                          SMS"}
     ]
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": false
   },
   "version": "privacyIDEA unknown"
 }

In this example two challenges are triggered, one with an email and one with an SMS. The application and thus the user has to decide, which one to use. They can use either.

Note

All challenge response tokens have the same transaction_id in this case.

GET /validate/samlcheck

Authenticate the user and return the SAML user information.

Parameters:

• user – The loginname/username of the user, who tries to authenticate.
• realm – The realm of the user, who tries to authenticate. If the realm is omitted, the user is looked up in the default realm.
• pass – The password, that consists of the OTP PIN and the OTP value.

Return:

a json result with a boolean “result”: true

Example response for a successful authentication:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "detail": {
     "message": "matching 1 tokens",
     "serial": "PISP0000AB00",
     "type": "spass"
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": {"attributes": {
                 "username": "koelbel",
                 "realm": "themis",
                 "mobile": null,
                 "phone": null,
                 "myOwn": "/data/file/home/koelbel",
                 "resolver": "themis",
                 "surname": "Kölbel",
                 "givenname": "Cornelius",
                 "email": null},
               "auth": true}
   },
   "version": "privacyIDEA unknown"
 }

The response in value->attributes can contain additional attributes (like “myOwn”) which you can define in the LDAP resolver in the attribute mapping.

POST /validate/samlcheck

Authenticate the user and return the SAML user information.

Parameters:

• user – The loginname/username of the user, who tries to authenticate.
• realm – The realm of the user, who tries to authenticate. If the realm is omitted, the user is looked up in the default realm.
• pass – The password, that consists of the OTP PIN and the OTP value.

Return:

a json result with a boolean “result”: true

Example response for a successful authentication:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "detail": {
     "message": "matching 1 tokens",
     "serial": "PISP0000AB00",
     "type": "spass"
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": {"attributes": {
                 "username": "koelbel",
                 "realm": "themis",
                 "mobile": null,
                 "phone": null,
                 "myOwn": "/data/file/home/koelbel",
                 "resolver": "themis",
                 "surname": "Kölbel",
                 "givenname": "Cornelius",
                 "email": null},
               "auth": true}
   },
   "version": "privacyIDEA unknown"
 }

The response in value->attributes can contain additional attributes (like “myOwn”) which you can define in the LDAP resolver in the attribute mapping.

GET /validate/check

check the authentication for a user or a serial number. Either a serial or a user is required to authenticate. The PIN and OTP value is sent in the parameter pass. In case of successful authentication it returns result->value: true.

In case /validate/radiuscheck is requested, the responses are modified as follows: A successful authentication returns an empty HTTP 204 response. An unsuccessful authentication returns an empty HTTP 400 response. Error responses are the same responses as for the /validate/check endpoint.

Parameters:

• serial – The serial number of the token, that tries to authenticate.
• user – The loginname/username of the user, who tries to authenticate.
• realm – The realm of the user, who tries to authenticate. If the realm is omitted, the user is looked up in the default realm.
• pass – The password, that consists of the OTP PIN and the OTP value.
• otponly – If set to 1, only the OTP value is verified. This is used in the management UI. Only used with the parameter serial.
• transaction_id – The transaction ID for a response to a challenge request
• state – The state ID for a response to a challenge request

Return:

a json result with a boolean “result”: true

Example Validation Request:

POST /validate/check HTTP/1.1
Host: example.com
Accept: application/json

user=user
realm=realm1
pass=s3cret123456

Example response for a successful authentication:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "detail": {
     "message": "matching 1 tokens",
     "serial": "PISP0000AB00",
     "type": "spass"
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": true
   },
   "version": "privacyIDEA unknown"
 }

Example response for this first part of a challenge response authentication:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "detail": {
     "serial": "PIEM0000AB00",
     "type": "email",
     "transaction_id": "12345678901234567890",
     "multi_challenge: [ {"serial": "PIEM0000AB00",
                          "transaction_id":  "12345678901234567890",
                          "message": "Please enter otp from your
                          email"},
                         {"serial": "PISM12345678",
                          "transaction_id": "12345678901234567890",
                          "message": "Please enter otp from your
                          SMS"}
     ]
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": false
   },
   "version": "privacyIDEA unknown"
 }

In this example two challenges are triggered, one with an email and one with an SMS. The application and thus the user has to decide, which one to use. They can use either.

Note

All challenge response tokens have the same transaction_id in this case.

POST /validate/check

check the authentication for a user or a serial number. Either a serial or a user is required to authenticate. The PIN and OTP value is sent in the parameter pass. In case of successful authentication it returns result->value: true.

In case /validate/radiuscheck is requested, the responses are modified as follows: A successful authentication returns an empty HTTP 204 response. An unsuccessful authentication returns an empty HTTP 400 response. Error responses are the same responses as for the /validate/check endpoint.

Parameters:

• serial – The serial number of the token, that tries to authenticate.
• user – The loginname/username of the user, who tries to authenticate.
• realm – The realm of the user, who tries to authenticate. If the realm is omitted, the user is looked up in the default realm.
• pass – The password, that consists of the OTP PIN and the OTP value.
• otponly – If set to 1, only the OTP value is verified. This is used in the management UI. Only used with the parameter serial.
• transaction_id – The transaction ID for a response to a challenge request
• state – The state ID for a response to a challenge request

Return:

a json result with a boolean “result”: true

Example Validation Request:

POST /validate/check HTTP/1.1
Host: example.com
Accept: application/json

user=user
realm=realm1
pass=s3cret123456

Example response for a successful authentication:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "detail": {
     "message": "matching 1 tokens",
     "serial": "PISP0000AB00",
     "type": "spass"
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": true
   },
   "version": "privacyIDEA unknown"
 }

Example response for this first part of a challenge response authentication:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "detail": {
     "serial": "PIEM0000AB00",
     "type": "email",
     "transaction_id": "12345678901234567890",
     "multi_challenge: [ {"serial": "PIEM0000AB00",
                          "transaction_id":  "12345678901234567890",
                          "message": "Please enter otp from your
                          email"},
                         {"serial": "PISM12345678",
                          "transaction_id": "12345678901234567890",
                          "message": "Please enter otp from your
                          SMS"}
     ]
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": false
   },
   "version": "privacyIDEA unknown"
 }

In this example two challenges are triggered, one with an email and one with an SMS. The application and thus the user has to decide, which one to use. They can use either.

Note

All challenge response tokens have the same transaction_id in this case.

System endpoints

This is the REST API for system calls to create and read system configuration.

The code of this module is tested in tests/test_api_system.py

GET /system/documentation

returns an restructured text document, that describes the complete configuration.

POST /system/setDefault

define default settings for tokens. These default settings are used when new tokens are generated. The default settings will not affect already enrolled tokens.

JSON Parameters:  

• DefaultMaxFailCount – Default value for the maximum allowed authentication failures
• DefaultSyncWindow – Default value for the synchronization window
• DefaultCountWindow – Default value for the counter window
• DefaultOtpLen – Default value for the OTP value length – usually 6 or 8
• DefaultResetFailCount – Default value, if the FailCounter should be reset on successful authentication [True|False]

Return:

a json result with a boolean “result”: true

POST /system/setConfig

set a configuration key or a set of configuration entries

parameter are generic keyname=value pairs.

remark In case of key-value pairs the type information could be

provided by an additional parameter with same keyname with the postfix ”.type”. Value could then be ‘password’ to trigger the storing of the value in an encrypted form

JSON Parameters:  

• key – configuration entry name
• value – configuration value
• type – type of the value: int or string/text or password. password will trigger to store the encrypted value
• description – additional information for this config entry

or

JSON Parameters:  

• pairs (key-value) – pair of &keyname=value pairs

Return:

a json result with a boolean “result”: true

Example request 1:

POST /system/setConfig
key=splitAtSign
value=true

Host: example.com
Accept: application/json

Example request 2:

POST /system/setConfig
BINDDN=myName
BINDPW=mySecretPassword
BINDPW.type=password

Host: example.com
Accept: application/json

GET /system/gpgkeys

Returns the GPG keys in the config directory specified by PI_GNUPG_HOME.

Return:A json list of the public GPG keys

GET /system/random

This endpoint can be used to retrieve random keys from privacyIDEA. In certain cases the client might need random data to initialize tokens on the client side. E.g. the command line client when initializing the yubikey or the WebUI when creating Client API keys for the yubikey.

In this case, privacyIDEA can created the random data/keys.

Query Parameters:  

• len – The length of a symmetric key (byte)
• encode – The type of encoding. Can be “hex” or “b64”.

Return:

key material

POST /system/hsm

Set the password for the security module

GET /system/hsm

Get the status of the security module.

GET /system/

This endpoint either returns all config entries or only the value of the one config key.

This endpoint can be called by the administrator but also by the normal user, so that the normal user gets necessary information about the system config

Parameters:

• key – The key to return.

Return:

A json response or a single value, when queried with a key.

Rtype:

json or scalar

POST /system/test/(tokentype)

The call /system/test/email tests the configuration of the email token.

GET /system/(key)

This endpoint either returns all config entries or only the value of the one config key.

This endpoint can be called by the administrator but also by the normal user, so that the normal user gets necessary information about the system config

Parameters:

• key – The key to return.

Return:

A json response or a single value, when queried with a key.

Rtype:

json or scalar

DELETE /system/(key)

delete a configuration key

JSON Parameters:  

• key – configuration key name

Returns:

a json result with the deleted value

Resolver endpoints

The code of this module is tested in tests/test_api_system.py

POST /resolver/test

Return:a json result with True, if the given values can create a working resolver and a description.

GET /resolver/

returns a json list of all resolver.

Parameters:

• type (basestring) – Only return resolvers of type (like passwdresolver..)
• editable (basestring) – Set to “1” if only editable resolvers should be returned.

POST /resolver/(resolver)

This creates a new resolver or updates an existing one. A resolver is uniquely identified by its name.

If you update a resolver, you do not need to provide all parameters. Parameters you do not provide are left untouched. When updating a resolver you must not change the type! You do not need to specify the type, but if you specify a wrong type, it will produce an error.

Parameters:

• resolver (basestring) – the name of the resolver.
• type – the type of the resolver. Valid types are passwdresolver,

ldapresolver, sqlresolver, scimresolver :type type: string :return: a json result with the value being the database id (>0)

Additional parameters depend on the resolver type.

LDAP:

• LDAPURI
• LDAPBASE
• BINDDN
• BINDPW
• TIMEOUT
• SIZELIMIT
• LOGINNAMEATTRIBUTE
• LDAPSEARCHFILTER
• LDAPFILTER
• USERINFO
• NOREFERRALS - True|False
• EDITABLE - True|False

SQL:

• Database
• Driver
• Server
• Port
• User
• Password
• Table
• Map

Passwd

• Filename

DELETE /resolver/(resolver)

This function deletes an existing resolver. A resolver can not be deleted, if it is contained in a realm

Parameters:

• resolver – the name of the resolver to delete.

Return:

json with success or fail

GET /resolver/(resolver)

This function retrieves the definition of a single resolver.

Parameters:

• resolver – the name of the resolver

Return:

a json result with the configuration of a specified resolver

Realm endpoints

The realm endpoints are used to define realms. A realm groups together many users. Administrators can manage the tokens of the users in such a realm. Policies and tokens can be assigned to realms.

A realm consists of several resolvers. Thus you can create a realm and gather users from LDAP and flat file source into one realm or you can pick resolvers that collect users from different points from your vast LDAP directory and group these users into a realm.

You will only be able to see and use user object, that are contained in a realm.

The code of this module is tested in tests/test_api_system.py

GET /realm/superuser

This call returns the list of all superuser realms as they are defined in pi.cfg. See The Config File for more information about this.

Return:a json result with a list of realms

Example request:

GET /superuser HTTP/1.1
Host: example.com
Accept: application/json

Example response:

HTTP/1.1 200 OK
Content-Type: application/json

{
  "id": 1,
  "jsonrpc": "2.0",
  "result": {
    "status": true,
    "value": ["superuser",
              "realm2"]
    }
  },
  "version": "privacyIDEA unknown"
}

GET /realm/

This call returns the list of all defined realms. It takes no arguments.

Return:a json result with a list of realms

Example request:

GET / HTTP/1.1
Host: example.com
Accept: application/json

Example response:

HTTP/1.1 200 OK
Content-Type: application/json

{
  "id": 1,
  "jsonrpc": "2.0",
  "result": {
    "status": true,
    "value": {
      "realm1_with_resolver": {
        "default": true,
        "resolver": [
          {
            "name": "reso1_with_realm",
            "type": "passwdresolver"
          }
        ]
      }
    }
  },
  "version": "privacyIDEA unknown"
}

POST /realm/(realm)

This call creates a new realm or reconfigures a realm. The realm contains a list of resolvers.

In the result it returns a list of added resolvers and a list of resolvers, that could not be added.

Parameters:

• realm – The unique name of the realm
• resolvers (string or list) – A comma separated list of unique resolver names or a list object
• priority – Additional parameters priority.<resolvername> define the priority of the resolvers within this realm.

Return:

a json result with a list of Realms

Example request:

To create a new realm “newrealm”, that consists of the resolvers “reso1_with_realm” and “reso2_with_realm” call:

POST /realm/newrealm HTTP/1.1
Host: example.com
Accept: application/json
Content-Length: 26
Content-Type: application/x-www-form-urlencoded

resolvers=reso1_with_realm, reso2_with_realm
priority.reso1_with_realm=1
priority.reso2_with_realm=2

Example response:

HTTP/1.1 200 OK
Content-Type: application/json

{
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
             "status": true,
             "value": {
                 "added": ["reso1_with_realm", "reso2_with_realm"],
                 "failed": []
             }
   }
   "version": "privacyIDEA unknown"
}

DELETE /realm/(realm)

This call deletes the given realm.

Parameters:

• realm – The name of the realm to delete

Return:

a json result with value=1 if deleting the realm was successful

Example request:

DELETE /realm/realm_to_delete HTTP/1.1
Host: example.com
Accept: application/json

Example response:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": 1
   },
   "version": "privacyIDEA unknown"
 }

Default Realm endpoints

These endpoints are used to define the default realm, retrieve it and delete it.

DELETE /defaultrealm

This call deletes the default realm.

Return:a json result with either 1 (success) or 0 (fail)

Example response:

 {
  "id": 1,
  "jsonrpc": "2.0",
  "result": {
    "status": true,
    "value": 1
  },
  "version": "privacyIDEA unknown"
}

GET /defaultrealm

This call returns the default realm

Return:a json description of the default realm with the resolvers

Example response:

{
  "id": 1,
  "jsonrpc": "2.0",
  "result": {
    "status": true,
    "value": {
      "defrealm": {
        "default": true,
        "resolver": [
          {
            "name": "defresolver",
            "type": "passwdresolver"
          }
        ]
      }
    }
  },
  "version": "privacyIDEA unknown"
}

POST /defaultrealm/(realm)

This call sets the default realm.

Parameters:

• realm – the name of the realm, that should be the default realm

Return:

a json result with either 1 (success) or 0 (fail)

Token endpoints

The token API can be accessed via /token.

You need to authenticate to gain access to these token functions. If you are authenticated as administrator, you can manage all tokens. If you are authenticated as normal user, you can only manage your own tokens. Some API calls are only allowed to be accessed by adminitrators.

To see how to authenticate read Authentication endpoints.

GET /token/challenges/

This endpoint returns the active challenges in the database or returns the challenges for a single token by its serial number

Query Parameters:  

• serial – The optional serial number of the token for which the challenges should be returned
• sortby – sort the output by column
• sortdir – asc/desc
• page – request a certain page
• pagesize – limit the number of returned tokens

Return:

json

POST /token/unassign

Unssign a token from a user. You can either provide “serial” as an argument to unassign this very token or you can provide user and realm, to unassign all tokens of a user.

Return:In case of success it returns “value”: True. Rtype:json object

POST /token/copyuser

Copy the token user from one token to the other.

JSON Parameters:  

• from (basestring) – the serial number of the token, from where you want to copy the pin.
• to (basestring) – the serial number of the token, from where you want to copy the pin.

Return:

returns value=True in case of success

Rtype:

bool

POST /token/disable

Disable a single token or all the tokens of a user either by providing the serial number of the single token or a username and realm.

Disabled tokens can not be used to authenticate but can be enabled again.

JSON Parameters:  

• serial (basestring) – the serial number of the single token to disable
• user (basestring) – The login name of the user
• realm (basestring) – the realm name of the user

Return:

In case of success it returns the number of disabled tokens in “value”.

Rtype:

json object

POST /token/copypin

Copy the token PIN from one token to the other.

JSON Parameters:  

• from (basestring) – the serial number of the token, from where you want to copy the pin.
• to (basestring) – the serial number of the token, from where you want to copy the pin.

Return:

returns value=True in case of success

Rtype:

bool

POST /token/assign

Assign a token to a user.

JSON Parameters:  

• serial – The token, which should be assigned to a user
• user – The username of the user
• realm – The realm of the user

Return:

In case of success it returns “value”: True.

Rtype:

json object

POST /token/revoke

Revoke a single token or all the tokens of a user. A revoked token will usually be locked. A locked token can not be used anymore. For certain token types additional actions might occur when revoking a token.

JSON Parameters:  

• serial (basestring) – the serial number of the single token to revoke
• user (basestring) – The login name of the user
• realm (basestring) – the realm name of the user

Return:

In case of success it returns the number of revoked tokens in “value”.

Rtype:

JSON object

POST /token/enable

Enable a single token or all the tokens of a user.

JSON Parameters:  

• serial (basestring) – the serial number of the single token to enable
• user (basestring) – The login name of the user
• realm (basestring) – the realm name of the user

Return:

In case of success it returns the number of enabled tokens in “value”.

Rtype:

json object

POST /token/resync

Resync the OTP token by providing two consecutive OTP values.

JSON Parameters:  

• serial (basestring) – the serial number of the single token to reset
• otp1 (basestring) – First OTP value
• otp2 (basestring) – Second OTP value

Return:

In case of success it returns “value”=True

Rtype:

json object

POST /token/setpin

Set the the user pin or the SO PIN of the specific token. Usually these are smartcard or token specific PINs. E.g. the userpin is used with mOTP tokens to store the mOTP PIN.

The token is identified by the unique serial number.

JSON Parameters:  

• serial (basestring) – the serial number of the single token to reset
• userpin (basestring) – The user PIN of a smartcard
• sopin (basestring) – The SO PIN of a smartcard
• otppin (basestring) – The OTP PIN of a token

Return:

In “value” returns the number of PINs set.

Rtype:

json object

POST /token/reset

Reset the failcounter of a single token or of all tokens of a user.

JSON Parameters:  

• serial (basestring) – the serial number of the single token to reset
• user (basestring) – The login name of the user
• realm (basestring) – the realm name of the user

Return:

In case of success it returns “value”=True

Rtype:

json object

POST /token/init

create a new token.

JSON Parameters:  

• otpkey – required: the secret key of the token
• genkey – set to =1, if key should be generated. We either need otpkey or genkey
• keysize – the size (byte) of the key. Either 20 or 32. Default is 20
• serial – the serial number/identifier of the token
• description – A description for the token
• pin – the pin of the token. “OTP PIN”
• user – the login user name. This user gets the token assigned
• realm – the realm of the user.
• type – the type of the token
• tokenrealm – additional realms, the token should be put into
• otplen – length of the OTP value
• hashlib – used hashlib sha1, sha256 or sha512
• validity_period_start – The beginning of the validity period
• validity_period_end – The end of the validity period
• 2stepinit – set to =1 in conjunction with genkey=1 if you want a 2 step initialization process. Additional policies have to be set see Two Step Enrollment.
• otpkeyformat – used to supply the OTP key in alternate formats, currently hex or base32check (see Two Step Enrollment)

Return:

a json result with a boolean “result”: true

Depending on the token type there can be additional parameters. In the tokenclass you can see additional parameters in the method update when looking for getParam functions.

Example response:

HTTP/1.1 200 OK
Content-Type: application/json

{
   "detail": {
     "googleurl": {
       "description": "URL for google Authenticator",
       "img": "<img    width=250   src="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAcIAAAHCAQAAAABUY/ToAAADsUlEQVR4nO2czY3bMBCF34QCfKSALcClyB2kpCAlpQOxlBQQgDwaoPBy4I+p9W4OSRaWF28OgizxgylgMJw/0oi/k/DlL0FApEiRIkWKFCnyeKRVmdrjNAFh3srTMuSS2qjLg2cr8pDkQpKMgF3SBITz1QA4YolVfQA4kiT35CNmK/JQZLM8aQaWH+3pEkEgTZlhBojksgGAAS7/83+K/ORkOF/NLtismiCfYXbOd+AxZivygCTXdCLCDJRLfTbhTo4wW5FHIJtyeAJIAJb4AobLBIP/ZQRAwMcyakxIPtd3ivw4EqObXJzody9t1EKS63N9p8iPI4sO3QTwGSSbA1Q0x+cWunWRDolsUjSnxvau6VB0xMIMrp4EPAnAkWsjpEMiu+ysD1mUZomuKk1/i6WtedIhkXupS1MEsMRmaVafh7dVfXwGV0D+kMj3yXDOsIsngXQiV59R0tZIE7jC0b4VA3WE2Yo8CtkTPy7b8sPA8HWbWML6dCKAqxG4GgADw+weOVuRRyTHuGztbk+PwdqQPIzTWibyDbJWVdOJQDLj9xkod4yOCK2gbzZvVpyip/xOkR9B4maCbnF8c53vHGuuLVaTHRLZpBgYgweAVP0hLPElA+mFtVrvf3W/aTM+brYij0j23o8JthAweNc1J5cCmSFNYDCAS5wfOVuRRyT7QpVL9F6XLN/zjhG4ZSAHj1trmcgmLcfoWoq6/B4LZLeqBxmVpxb5WobYfl8vaxfU7DSA4mdLh0S+TW5W2xXTiaWZ0WbALqiXmi5KU/n5tN8p8r+TzaqUH936MKNW6/2uIkvZIZF/IEleDfAZZnYi1zSB/DmVpa2YJZtVLxP5JmnfWCutty5qwNcFrWSsV2xGxs3+03+K/Cxk74WtTWflDr652L0XtoZuylOLvJNb9H7XPzQ0DOX9RTokcpAhAzRYpN4LO5TsI1rQLx0SOci4z7VcSuvQZgxWX1gfbfBX1ctEvhLupbZSe5bNQK0Jv/dTe9U6RL6WtoIBqDs33NA7Xdey3SYzrWUi99L8IfJW4cC4pYNjg+Ow/+O5vlPkx5OpnSsUzler2cbS29g8pmBmWH6elGMU+UqaFwS0NBBa9O45Rmhr26Mof0jkTt440MNlC9aOGQqzA8McaQs34xJfsv3rf4r8XOTduR+lezHN5fyh0sdY76qz/cDZijwwGcxqs0c9gNFx5w9t7e18hNmKPBRZ7NDtXKF6V1qp2e9qtZ7DkOf6TpEiRYoUKVKkyPfkNyq7YXtdjZCIAAAAAElFTkSuQmCC"/>",
       "value": "otpauth://hotp/mylabel?secret=GEZDGNBVGY3TQOJQGEZDGNBVGY3TQOJQ&counter=0"
     },
     "oathurl": {
       "description": "URL for OATH token",
       "img": "<img    width=250   src="data:image/png;base64,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"/>",
       "value": "oathtoken:///addToken?name=mylabel&lockdown=true&key=3132333435363738393031323334353637383930"
     },
     "otpkey": {
       "description": "OTP seed",
       "img": "<img    width=200   src="data:image/png;base64,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"/>",
       "value": "seed://3132333435363738393031323334353637383930"
     },
     "serial": "OATH00096020"
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": true
   },
   "version": "privacyIDEA unknown"
 }

2 Step Enrollment

Some tokens might need a 2 step initialization process like a smartphone app. This way you can create a shared secret from a part generated by the privacyIDEA server and from a second part generated by the smartphone app/client.

The first API call would be

POST /token/init

2stepinit=1

The response would contain the otpkey generated by the server and the serial number of the token. At this point, the token is deactivated and marked as being in an enrollment state. The client would also generated a component of the key and send his component to the privacyIDEA server:

The second API call would be

POST /token/init

serial=<serial from the previous response>
otpkey=<key part generated by the client>

Each tokenclass can define its own way to generate the secret key by overwriting the method generate_symmetric_key. The Base Tokenclass contains an extremely simple way by concatenating the two parts. See generate_symmetric_key()

POST /token/set

This API is only to be used by the admin! This can be used to set token specific attributes like

• description
• count_window
• sync_window
• count_auth_max
• count_auth_success_max
• hashlib,
• max_failcount
• validity_period_start
• validity_period_end

The token is identified by the unique serial number or by the token owner. In the later case all tokens of the owner will be modified.

The validity period needs to be provided in the format YYYY-MM-DDThh:mm+oooo

JSON Parameters:  

• serial (basestring) – the serial number of the single token to reset
• user (basestring) – The username of the token owner
• realm (basestring) – The realm name of the token owner

Return:

returns the number of attributes set in “value”

Rtype:

json object

GET /token/

Display the list of tokens. Using different parameters you can choose, which tokens you want to get and also in which format you want to get the information (outform).

Query Parameters:  

• serial – Display the token data of this single token. You can do a not strict matching by specifying a serial like “OATH”.
• type – Display only token of type. You ca do a non strict matching by specifying a tokentype like “otp”, to file hotp and totp tokens.
• user – display tokens of this user
• tokenrealm – takes a realm, only the tokens in this realm will be displayed
• description (basestring) – Display token with this kind of description
• sortby – sort the output by column
• sortdir – asc/desc
• page – request a certain page
• assigned – Only return assigned (True) or not assigned (False) tokens
• pagesize – limit the number of returned tokens
• user_fields – additional user fields from the userid resolver of the owner (user)
• outform – if set to “csv”, than the token list will be given in CSV

Return:

a json result with the data being a list of token dictionaries:

{ "data": [ { <token1> }, { <token2> } ]}

Rtype:

json

POST /token/info/(serial)/(key)

Add a specific tokeninfo entry to a token. Already existing entries with the same key are overwritten.

Parameters:

• serial – the serial number/identifier of the token
• key – token info key that should be set

Query Parameters:  

• value – token info value that should be set

Return:

returns value=True in case the token info could be set

Rtype:

bool

DELETE /token/info/(serial)/(key)

Delete a specific tokeninfo entry of a token.

Parameters:

• serial – the serial number/identifier of the token
• key – token info key that should be deleted

Return:

returns value=True in case a matching token was found, which does not necessarily mean

that the matching token had a tokeninfo value set in the first place. :rtype: bool

GET /token/challenges/(serial)

This endpoint returns the active challenges in the database or returns the challenges for a single token by its serial number

Query Parameters:  

• serial – The optional serial number of the token for which the challenges should be returned
• sortby – sort the output by column
• sortdir – asc/desc
• page – request a certain page
• pagesize – limit the number of returned tokens

Return:

json

GET /token/getserial/(otp)

Get the serial number for a given OTP value. If the administrator has a token, he does not know to whom it belongs, he can type in the OTP value and gets the serial number of the token, that generates this very OTP value.

Query Parameters:  

• otp – The given OTP value
• type – Limit the search to this token type
• unassigned – If set=1, only search in unassigned tokens
• assigned – If set=1, only search in assigned tokens
• count – if set=1, only return the number of tokens, that will be searched
• serial – This can be a substring of serial numbers to search in.
• window – The number of OTP look ahead (default=10)

Return:

The serial number of the token found

POST /token/disable/(serial)

Disable a single token or all the tokens of a user either by providing the serial number of the single token or a username and realm.

Disabled tokens can not be used to authenticate but can be enabled again.

JSON Parameters:  

• serial (basestring) – the serial number of the single token to disable
• user (basestring) – The login name of the user
• realm (basestring) – the realm name of the user

Return:

In case of success it returns the number of disabled tokens in “value”.

Rtype:

json object

POST /token/revoke/(serial)

Revoke a single token or all the tokens of a user. A revoked token will usually be locked. A locked token can not be used anymore. For certain token types additional actions might occur when revoking a token.

JSON Parameters:  

• serial (basestring) – the serial number of the single token to revoke
• user (basestring) – The login name of the user
• realm (basestring) – the realm name of the user

Return:

In case of success it returns the number of revoked tokens in “value”.

Rtype:

JSON object

POST /token/enable/(serial)

Enable a single token or all the tokens of a user.

JSON Parameters:  

• serial (basestring) – the serial number of the single token to enable
• user (basestring) – The login name of the user
• realm (basestring) – the realm name of the user

Return:

In case of success it returns the number of enabled tokens in “value”.

Rtype:

json object

POST /token/resync/(serial)

Resync the OTP token by providing two consecutive OTP values.

JSON Parameters:  

• serial (basestring) – the serial number of the single token to reset
• otp1 (basestring) – First OTP value
• otp2 (basestring) – Second OTP value

Return:

In case of success it returns “value”=True

Rtype:

json object

POST /token/setpin/(serial)

Set the the user pin or the SO PIN of the specific token. Usually these are smartcard or token specific PINs. E.g. the userpin is used with mOTP tokens to store the mOTP PIN.

The token is identified by the unique serial number.

JSON Parameters:  

• serial (basestring) – the serial number of the single token to reset
• userpin (basestring) – The user PIN of a smartcard
• sopin (basestring) – The SO PIN of a smartcard
• otppin (basestring) – The OTP PIN of a token

Return:

In “value” returns the number of PINs set.

Rtype:

json object

POST /token/reset/(serial)

Reset the failcounter of a single token or of all tokens of a user.

JSON Parameters:  

• serial (basestring) – the serial number of the single token to reset
• user (basestring) – The login name of the user
• realm (basestring) – the realm name of the user

Return:

In case of success it returns “value”=True

Rtype:

json object

POST /token/realm/(serial)

Set the realms of a token. The token is identified by the unique serial number

You can call the function like this:

POST /token/realm?serial=<serial>&realms=<something> POST /token/realm/<serial>?realms=<hash>

JSON Parameters:  

• serial (basestring) – the serial number of the single token to reset
• realms (basestring) – The realms the token should be assigned to. Comma separated

Return:

returns value=True in case of success

Rtype:

bool

POST /token/load/(filename)

The call imports the given file containing token definitions. The file can be an OATH CSV file, an aladdin XML file or a Yubikey CSV file exported from the yubikey initialization tool.

The function is called as a POST request with the file upload.

JSON Parameters:  

• filename – The name of the token file, that is imported
• type – The file type. Can be “aladdin-xml”, “oathcsv” or “yubikeycsv”.
• tokenrealms – comma separated list of tokens.
• psk – Pre Shared Key, when importing PSKC

Return:

The number of the imported tokens

Rtype:

int

POST /token/lost/(serial)

Mark the specified token as lost and create a new temporary token. This new token gets the new serial number “lost<old-serial>” and a certain validity period and the PIN of the lost token.

This method can be called by either the admin or the user on his own tokens.

You can call the function like this:

POST /token/lost/serial

JSON Parameters:  

• serial (basestring) – the serial number of the lost token.

Return:

returns value=dictionary in case of success

Rtype:

bool

POST /token/set/(serial)

This API is only to be used by the admin! This can be used to set token specific attributes like

• description
• count_window
• sync_window
• count_auth_max
• count_auth_success_max
• hashlib,
• max_failcount
• validity_period_start
• validity_period_end

The token is identified by the unique serial number or by the token owner. In the later case all tokens of the owner will be modified.

The validity period needs to be provided in the format YYYY-MM-DDThh:mm+oooo

JSON Parameters:  

• serial (basestring) – the serial number of the single token to reset
• user (basestring) – The username of the token owner
• realm (basestring) – The realm name of the token owner

Return:

returns the number of attributes set in “value”

Rtype:

json object

DELETE /token/(serial)

Delete a token by its serial number or delete all tokens of a user.

JSON Parameters:  

• serial – The serial number of a single token.
• user – The username of the user, whose tokens should be deleted.
• realm – The realm of the user.

Return:

In case of success it return the number of deleted tokens in “value”

Rtype:

json object

User endpoints

The user endpoints is a subset of the system endpoint.

GET /user/

list the users in a realm

A normal user can call this endpoint and will get information about his own account.

Parameters:

• realm – a realm that contains several resolvers. Only show users from this realm
• resolver – a distinct resolvername
• <searchexpr> – a search expression, that depends on the ResolverClass

Return:

json result with “result”: true and the userlist in “value”.

Example request:

GET /user?realm=realm1 HTTP/1.1
Host: example.com
Accept: application/json

Example response:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": [
       {
         "description": "Cornelius K\u00f6lbel,,+49 151 2960 1417,+49 561 3166797,cornelius.koelbel@netknights.it",
         "email": "cornelius.koelbel@netknights.it",
         "givenname": "Cornelius",
         "mobile": "+49 151 2960 1417",
         "phone": "+49 561 3166797",
         "surname": "K\u00f6lbel",
         "userid": "1009",
         "username": "cornelius",
         "resolver": "name-of-resolver"
       }
     ]
   },
   "version": "privacyIDEA unknown"
 }

POST /user/

Create a new user in the given resolver.

Example request:

POST /user
user=new_user
resolver=<resolvername>
surname=...
givenname=...
email=...
mobile=...
phone=...
password=...
description=...

Host: example.com
Accept: application/json

POST /user

Create a new user in the given resolver.

Example request:

POST /user
user=new_user
resolver=<resolvername>
surname=...
givenname=...
email=...
mobile=...
phone=...
password=...
description=...

Host: example.com
Accept: application/json

PUT /user/

Edit a user in the user store. The resolver must have the flag editable, so that the user can be deleted. Only administrators are allowed to edit users.

Example request:

PUT /user
user=existing_user
resolver=<resolvername>
surname=...
givenname=...
email=...
mobile=...
phone=...
password=...
description=...

Host: example.com
Accept: application/json

Note

Also a user can call this function to e.g. change his password. But in this case the parameter “user” and “resolver” get overwritten by the values of the authenticated user, even if he specifies another username.

PUT /user

Edit a user in the user store. The resolver must have the flag editable, so that the user can be deleted. Only administrators are allowed to edit users.

Example request:

PUT /user
user=existing_user
resolver=<resolvername>
surname=...
givenname=...
email=...
mobile=...
phone=...
password=...
description=...

Host: example.com
Accept: application/json

Note

Also a user can call this function to e.g. change his password. But in this case the parameter “user” and “resolver” get overwritten by the values of the authenticated user, even if he specifies another username.

DELETE /user/(resolvername)/(username)

Delete a User in the user store. The resolver must have the flag editable, so that the user can be deleted. Only administrators are allowed to delete users.

Delete a user object in a user store by calling

Example request:

DELETE /user/<resolvername>/<username>
Host: example.com
Accept: application/json

The code of this module is tested in tests/test_api_system.py

Policy endpoints

The policy endpoints are a subset of the system endpoint.

You can read more about policies at Policies.

GET /policy/check

This function checks, if the given parameters would match a defined policy or not.

Query Parameters:  

• user – the name of the user
• realm – the realm of the user or the realm the administrator want to do administrative tasks on.
• resolver – the resolver of a user
• scope – the scope of the policy
• action – the action that is done - if applicable
• client (IP_Address) – the client, from which this request would be issued

Return:

a json result with the keys allowed and policy in the value key

Rtype:

json

Status Codes:

• 200 OK – Policy created or modified.
• 401 Unauthorized – Authentication failed

Example request:

GET /policy/check?user=admin&realm=r1&client=172.16.1.1 HTTP/1.1
Host: example.com
Accept: application/json

Example response:

HTTP/1.0 200 OK
Content-Type: application/json

 {
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": {
       "pol_update_del": {
         "action": "enroll",
         "active": true,
         "client": "172.16.0.0/16",
         "name": "pol_update_del",
         "realm": "r1",
         "resolver": "test",
         "scope": "selfservice",
         "time": "",
         "user": "admin"
       }
     }
   },
   "version": "privacyIDEA unknown"
 }

GET /policy/defs

This is a helper function that returns the POSSIBLE policy definitions, that can be used to define your policies.

Query Parameters:  

• scope – if given, the function will only return policy definitions for the given scope.

Return:

The policy definitions of the allowed scope with the actions and action types. The top level key is the scope.

Rtype:

dict

GET /policy/

this function is used to retrieve the policies that you defined. It can also be used to export the policy to a file.

Query Parameters:  

• name – will only return the policy with the given name
• export – The filename needs to be specified as the third part of the URL like policy.cfg. It will then be exported to this file.
• realm – will return all policies in the given realm
• scope – will only return the policies within the given scope
• active – Set to true or false if you only want to display active or inactive policies.

Return:

a json result with the configuration of the specified policies

Rtype:

json

Status Codes:

• 200 OK – Policy created or modified.
• 401 Unauthorized – Authentication failed

Example request:

In this example a policy “pol1” is created.

GET /policy/pol1 HTTP/1.1
Host: example.com
Accept: application/json

Example response:

HTTP/1.0 200 OK
Content-Type: application/json

 {
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": {
       "pol_update_del": {
         "action": "enroll",
         "active": true,
         "client": "1.1.1.1",
         "name": "pol_update_del",
         "realm": "r1",
         "resolver": "test",
         "scope": "selfservice",
         "time": "",
         "user": "admin"
       }
     }
   },
   "version": "privacyIDEA unknown"
 }

POST /policy/disable/(name)

Disable a given policy by its name.

JSON Parameters:  

• name – The name of the policy

Return:

ID in the database

POST /policy/enable/(name)

Enable a given policy by its name.

JSON Parameters:  

• name – Name of the policy

Return:

ID in the database

GET /policy/export/(export)

this function is used to retrieve the policies that you defined. It can also be used to export the policy to a file.

Query Parameters:  

• name – will only return the policy with the given name
• export – The filename needs to be specified as the third part of the URL like policy.cfg. It will then be exported to this file.
• realm – will return all policies in the given realm
• scope – will only return the policies within the given scope
• active – Set to true or false if you only want to display active or inactive policies.

Return:

a json result with the configuration of the specified policies

Rtype:

json

Status Codes:

• 200 OK – Policy created or modified.
• 401 Unauthorized – Authentication failed

Example request:

In this example a policy “pol1” is created.

GET /policy/pol1 HTTP/1.1
Host: example.com
Accept: application/json

Example response:

HTTP/1.0 200 OK
Content-Type: application/json

 {
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": {
       "pol_update_del": {
         "action": "enroll",
         "active": true,
         "client": "1.1.1.1",
         "name": "pol_update_del",
         "realm": "r1",
         "resolver": "test",
         "scope": "selfservice",
         "time": "",
         "user": "admin"
       }
     }
   },
   "version": "privacyIDEA unknown"
 }

POST /policy/import/(filename)

This function is used to import policies from a file.

JSON Parameters:  

• filename – The name of the file in the request

Form Parameters:  

• file – The uploaded file contents

Return:

A json response with the number of imported policies.

Status Codes:

• 200 OK – Policy created or modified.
• 401 Unauthorized – Authentication failed

Example request:

POST /policy/import/backup-policy.cfg HTTP/1.1
Host: example.com
Accept: application/json

Example response:

HTTP/1.0 200 OK
Content-Type: application/json

 {
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": 2
   },
   "version": "privacyIDEA unknown"
 }

GET /policy/defs/(scope)

This is a helper function that returns the POSSIBLE policy definitions, that can be used to define your policies.

Query Parameters:  

• scope – if given, the function will only return policy definitions for the given scope.

Return:

The policy definitions of the allowed scope with the actions and action types. The top level key is the scope.

Rtype:

dict

POST /policy/(name)

Creates a new policy that defines access or behaviour of different actions in privacyIDEA

JSON Parameters:  

• name (basestring) – name of the policy
• scope – the scope of the policy like “admin”, “system”, “authentication” or “selfservice”
• adminrealm – Realm of the administrator. (only for admin scope)
• action – which action may be executed
• realm – For which realm this policy is valid
• resolver – This policy is valid for this resolver
• user – The policy is valid for these users. string with wild cards or list of strings
• time – on which time does this policy hold
• client (IP address with subnet) – for which requesting client this should be
• active – bool, whether this policy is active or not
• check_all_resolvers – bool, whether all all resolvers in which the user exists should be checked with this policy.

Return:

a json result with success or error

Status Codes:

• 200 OK – Policy created or modified.
• 401 Unauthorized – Authentication failed

Example request:

In this example a policy “pol1” is created.

POST /policy/pol1 HTTP/1.1
Host: example.com
Accept: application/json

scope=admin
realm=realm1
action=enroll, disable

Example response:

HTTP/1.0 200 OK
Content-Length: 354
Content-Type: application/json

 {
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": {
       "setPolicy pol1": 1
     }
   },
   "version": "privacyIDEA unknown"
 }

GET /policy/(name)

this function is used to retrieve the policies that you defined. It can also be used to export the policy to a file.

Query Parameters:  

• name – will only return the policy with the given name
• export – The filename needs to be specified as the third part of the URL like policy.cfg. It will then be exported to this file.
• realm – will return all policies in the given realm
• scope – will only return the policies within the given scope
• active – Set to true or false if you only want to display active or inactive policies.

Return:

a json result with the configuration of the specified policies

Rtype:

json

Status Codes:

• 200 OK – Policy created or modified.
• 401 Unauthorized – Authentication failed

Example request:

In this example a policy “pol1” is created.

GET /policy/pol1 HTTP/1.1
Host: example.com
Accept: application/json

Example response:

HTTP/1.0 200 OK
Content-Type: application/json

 {
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": {
       "pol_update_del": {
         "action": "enroll",
         "active": true,
         "client": "1.1.1.1",
         "name": "pol_update_del",
         "realm": "r1",
         "resolver": "test",
         "scope": "selfservice",
         "time": "",
         "user": "admin"
       }
     }
   },
   "version": "privacyIDEA unknown"
 }

DELETE /policy/(name)

This deletes the policy of the given name.

JSON Parameters:  

• name – the policy with the given name

Return:

a json result about the delete success. In case of success value > 0

Status Codes:

• 200 OK – Policy created or modified.
• 401 Unauthorized – Authentication failed

Example request:

In this example a policy “pol1” is created.

DELETE /policy/pol1 HTTP/1.1
Host: example.com
Accept: application/json

Example response:

HTTP/1.0 200 OK
Content-Type: application/json

{
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": 1
   },
   "version": "privacyIDEA unknown"
}

This endpoint is used to create, modify, list and delete Machine Resolvers. Machine Resolvers fetch machine information from remote machine stores like a hosts file or an Active Directory.

The code of this module is tested in tests/test_api_machineresolver.py

Machine Resolver endpoints

POST /machineresolver/test

This function tests, if the given parameter will create a working machine resolver. The Machine Resolver Class itself verifies the functionality. This can also be network connectivity to a Machine Store.

Return:a json result with bool

GET /machineresolver/

returns a json list of all machine resolver.

Parameters:

• type – Only return resolvers of type (like “hosts”...)

POST /machineresolver/(resolver)

This creates a new machine resolver or updates an existing one. A resolver is uniquely identified by its name.

If you update a resolver, you do not need to provide all parameters. Parameters you do not provide are left untouched. When updating a resolver you must not change the type! You do not need to specify the type, but if you specify a wrong type, it will produce an error.

Parameters:

• resolver (basestring) – the name of the resolver.
• type (string) – the type of the resolver. Valid types are... “hosts”

Return:

a json result with the value being the database id (>0)

Additional parameters depend on the resolver type.

hosts:

• filename

DELETE /machineresolver/(resolver)

this function deletes an existing machine resolver

Parameters:

• resolver – the name of the resolver to delete.

Return:

json with success or fail

GET /machineresolver/(resolver)

This function retrieves the definition of a single machine resolver.

Parameters:

• resolver – the name of the resolver

Return:

a json result with the configuration of a specified resolver

This REST API is used to list machines from Machine Resolvers.

The code is tested in tests/test_api_machines

Machine endpoints

POST /machine/tokenoption

This sets a Machine Token option or deletes it, if the value is empty.

Parameters:

• hostname – identify the machine by the hostname
• machineid – identify the machine by the machine ID and the resolver name
• resolver – identify the machine by the machine ID and the resolver name
• serial – identify the token by the serial number
• application – the name of the application like “luks” or “ssh”.

Parameters not listed will be treated as additional options.

Return:

GET /machine/authitem

This fetches the authentication items for a given application and the given client machine.

Parameters:

• challenge (basestring) – A challenge for which the authentication item is calculated. In case of the Yubikey this can be a challenge that produces a response. The authentication item is the combination of the challenge and the response.
• hostname (basestring) – The hostname of the machine

Return:

dictionary with lists of authentication items

Example response:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": { "ssh": [ { "username": "....",
                           "sshkey": "...."
                         }
                       ],
                "luks": [ { "slot": ".....",
                            "challenge": "...",
                            "response": "...",
                            "partition": "..."
                        ]
              }
   },
   "version": "privacyIDEA unknown"
 }

POST /machine/token

Attach an existing token to a machine with a certain application.

Parameters:

• hostname – identify the machine by the hostname
• machineid – identify the machine by the machine ID and the resolver name
• resolver – identify the machine by the machine ID and the resolver name
• serial – identify the token by the serial number
• application – the name of the application like “luks” or “ssh”.

Parameters not listed will be treated as additional options.

Return:json result with “result”: true and the machine list in “value”.

Example request:

POST /token HTTP/1.1
Host: example.com
Accept: application/json

{ "hostname": "puckel.example.com",
  "machienid": "12313098",
  "resolver": "machineresolver1",
  "serial": "tok123",
  "application": "luks" }

GET /machine/token

Return a list of MachineTokens either for a given machine or for a given token.

Parameters:

• serial – Return the MachineTokens for a the given Token
• hostname – Identify the machine by the hostname
• machineid – Identify the machine by the machine ID and the resolver name
• resolver – Identify the machine by the machine ID and the resolver name

Return:

GET /machine/

List all machines that can be found in the machine resolvers.

Parameters:

• hostname – only show machines, that match this hostname as substring
• ip – only show machines, that exactly match this IP address
• id – filter for substring matching ids
• resolver – filter for substring matching resolvers
• any – filter for a substring either matching in “hostname”, “ip” or “id”

Return:

json result with “result”: true and the machine list in “value”.

Example request:

GET /hostname?hostname=on HTTP/1.1
Host: example.com
Accept: application/json

Example response:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": [
       {
         "id": "908asljdas90ad0",
         "hostname": [ "flavon.example.com", "test.example.com" ],
         "ip": "1.2.3.4",
         "resolver_name": "machineresolver1"
       },
       {
         "id": "1908209x48x2183",
         "hostname": [ "london.example.com" ],
         "ip": "2.4.5.6",
         "resolver_name": "machineresolver1"
       }
     ]
   },
   "version": "privacyIDEA unknown"
 }

DELETE /machine/token/(serial)/(machineid)/(resolver)/(application)

Detach a token from a machine with a certain application.

Parameters:

• machineid – identify the machine by the machine ID and the resolver name
• resolver – identify the machine by the machine ID and the resolver name
• serial – identify the token by the serial number
• application – the name of the application like “luks” or “ssh”.

Return:

json result with “result”: true and the machine list in “value”.

Example request:

DELETE /token HTTP/1.1
Host: example.com
Accept: application/json

{ "hostname": "puckel.example.com",
  "resolver": "machineresolver1",
  "application": "luks" }

GET /machine/authitem/(application)

This fetches the authentication items for a given application and the given client machine.

Parameters:

• challenge (basestring) – A challenge for which the authentication item is calculated. In case of the Yubikey this can be a challenge that produces a response. The authentication item is the combination of the challenge and the response.
• hostname (basestring) – The hostname of the machine

Return:

dictionary with lists of authentication items

Example response:

HTTP/1.1 200 OK
Content-Type: application/json

 {
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": { "ssh": [ { "username": "....",
                           "sshkey": "...."
                         }
                       ],
                "luks": [ { "slot": ".....",
                            "challenge": "...",
                            "response": "...",
                            "partition": "..."
                        ]
              }
   },
   "version": "privacyIDEA unknown"
 }

This endpoint is used to get the information from the server, which application types are known and which options these applications provide.

Applications are used to attach tokens to machines.

The code of this module is tested in tests/test_api_applications.py

Application endpoints

GET /application/

returns a json list of the available applications

Tokentype endpoints

This API endpoint is a generic endpoint that can be used by any token type.

The tokentype needs to implement a classmethod api_endpoint and can then be called by /ttype/<tokentype>. This way, each tokentype can create its own API without the need to change the core API.

The TiQR Token uses this API to implement its special functionalities. See TiQR Token.

GET /ttype/(ttype)

This is a special token function. Each token type can define an additional API call, that does not need authentication on the REST API level.

Return:Token Type dependent

POST /ttype/(ttype)

This is a special token function. Each token type can define an additional API call, that does not need authentication on the REST API level.

Return:Token Type dependent

SMTP server endpoints

This endpoint is used to create, update, list and delete SMTP server definitions. SMTP server definitions can be used for several purposes like EMail-Token, SMS Token with SMTP gateway, notification like PIN handler and registration.

The code of this module is tested in tests/test_api_smtpserver.py

POST /smtpserver/send_test_email

Test the email configuration :return:

GET /smtpserver/

This call gets the list of SMTP server definitions

POST /smtpserver/(identifier)

This call creates or updates an SMTP server definition.

Parameters:

• identifier – The unique name of the SMTP server definition
• server – The FQDN or IP of the mail server
• port – The port of the mail server
• username – The mail username for authentication at the SMTP server
• password – The password for authentication at the SMTP server
• tls – If the server should do TLS
• description – A description for the definition

DELETE /smtpserver/(identifier)

This call deletes the specified SMTP server configuration

Parameters:

• identifier – The unique name of the SMTP server definition

LIB level

At the LIB level all library functions are defined. There is no authentication on this level. Also there is no flask/Web/request code on this level.

Request information and the logged_in_user need to be passed to the functions as parameters, if they are needed.

If possible, policies are checked with policy decorators.

library functions

Based on the database models, which are tested in tests/test_db_model.py, there are different modules.

resolver.py contains functions to simply deal with resolver definitions. On this level users and realms are not know, yet.

realm.py contains functions to deal with realm. Realms are a list of several resolvers. So prior to bother the realm.py, the resolver.py should be understood and working. On this level, users are not known, yet.

user.py contains functions to deal with users. A user object is an entity in a realm. And of course the user object itself can be found in a resolver. But you need to have working resolver.py and realm.py to be able to work with user.py

For further details see the following modules:

Users

There are the library functions for user functions. It depends on the lib.resolver and lib.realm.

There are and must be no dependencies to the token functions (lib.token) or to webservices!

This code is tested in tests/test_lib_user.py

class privacyidea.lib.user.User(login='', realm='', resolver='')

The user has the attributes

login, realm and resolver.

Usually a user can be found via “login@realm”.

A user object with an empty login and realm should not exist, whereas a user object could have an empty resolver.

check_password(password)

The password of the user is checked against the user source

Parameters:password – The clear text password Returns:the username of the authenticated user. If unsuccessful, returns None Return type:string/None

delete()

This deletes the user in the user store. I.e. the user in the SQL database or the LDAP gets deleted.

Returns True in case of success

exist()

Check if the user object exists in the user store :return: True or False

get_ordererd_resolvers()

returns a list of resolvernames ordered by priority. The resolver with the lowest priority is the first. If resolvers have the same priority, they are ordered alphabetically.

Returns:list or resolvernames

get_search_fields()

Return the valid search fields of a user. The search fields are defined in the UserIdResolver class.

Returns:searchFields with name (key) and type (value) Return type:dict

get_user_identifiers()

This returns the UserId information from the resolver object and the resolvertype and the resolvername (former: getUserId) (former: getUserResolverId) :return: The userid, the resolver type and the resolver name

like (1000, “passwdresolver”, “resolver1”)

Return type:tuple

get_user_phone(phone_type='phone')

Returns the phone number of a user

Parameters:phone_type (string) – The type of the phone, i.e. either mobile or phone (land line) Returns:list with phone numbers of this user object

get_user_realms()

Returns a list of the realms, a user belongs to. Usually this will only be one realm. But if the user object has no realm but only a resolver, than all realms, containing this resolver are returned. This function is used for the policy module

Returns:realms of the user Return type:list

info

return the detailed information for the user

Returns:a dict with all the userinformation Return type:dict

is_empty()

login = ''

realm = ''

resolver = ''

update_user_info(attributes, password=None)

This updates the given attributes of a user. The attributes can be “username”, “surname”, “givenname”, “email”, “mobile”, “phone”, “password”

Parameters:

• attributes (dict) – A dictionary of the attributes to be updated
• password – The password of the user

Returns:

True in case of success

privacyidea.lib.user.create_user(resolvername, attributes, password=None)

This creates a new user in the given resolver. The resolver must be editable to do so.

The attributes is a dictionary containing the keys “username”, “email”, “phone”, “mobile”, “surname”, “givenname”, “password”.

We return the UID and not the user object, since the user could be located in several realms!

Parameters:

• resolvername (basestring) – The name of the resolver, in which the user should be created
• attributes (dict) – Attributes of the user
• password – The password of the user

Returns:

The uid of the user object

privacyidea.lib.user.get_user_from_param(param, optionalOrRequired=True)

Find the parameters user, realm and resolver and create a user object from these parameters.

An exception is raised, if a user in a realm is found in more than one resolvers.

Parameters:param (dict) – The dictionary of request parameters Returns:User as found in the parameters Return type:User object

privacyidea.lib.user.get_user_info(userid, resolvername)

return the detailed information for a user in a resolver

Parameters:

• userid (string) – The id of the user in a resolver
• resolvername – The name of the resolver

Returns:

a dict with all the userinformation

Return type:

dict

privacyidea.lib.user.get_user_list(param=None, user=None)

privacyidea.lib.user.get_username(userid, resolvername)

Determine the username for a given id and a resolvername.

Parameters:

• userid (string) – The id of the user in a resolver
• resolvername – The name of the resolver

Returns:

the username or “” if it does not exist

Return type:

string

privacyidea.lib.user.split_user(username)

Split the username of the form user@realm into the username and the realm splitting myemail@emailprovider.com@realm is also possible and will return (myemail@emailprovider, realm).

If for a user@domain the “domain” does not exist as realm, the name is not split, since it might be the user@domain in the default realm

We can also split realmuser to (user, realm)

Parameters:username (string) – the username to split Returns:username and realm Return type:tuple

Token Class

The following token types are known to privacyIDEA. All are inherited from the base tokenclass describe below.

4 Eyes Token

class privacyidea.lib.tokens.foureyestoken.FourEyesTokenClass(db_token)

The FourEyes token can be used to implement the Two Man Rule. The FourEyes token defines how many tokens of which realms are required like: * 2 tokens of RealmA * 1 token of RealmB

Then users (the owners of those tokens) need to login by everyone entering their OTP PIN and OTP value. It does not matter, in which order they enter the values. All their PINs and OTPs are concatenated into one password field but need to be separated by the splitting sign.

The FourEyes token again splits the password value and tries to authenticate each of the these passwords in the realms using the function check_realm_pass.

The FourEyes token itself does not provide an OTP PIN.

The token is initialized using additional parameters at token/init:

Example Authentication Request:

POST /auth HTTP/1.1
Host: example.com
Accept: application/json

type=4eyes
user=cornelius
realm=realm1
4eyes=realm1:2,realm2:1
separator=%20

authenticate(passw, user=None, options=None)

do the authentication on base of password / otp and user and options, the request parameters.

Here we contact the other privacyIDEA server to validate the OtpVal.

Parameters:

• passw – the password / otp
• user – the requesting user
• options – the additional request parameters

Returns:

tuple of (success, otp_count - 0 or -1, reply)

static convert_realms(realms)

This function converts the realms as given by the API parameter to a dictionary.

realm1:2,realm2:1 -> {“realm1”:2,

“realm2”:1}

Parameters:realms (basestring) – a serialized list of realms Returns:dict of realms

static get_class_info(key=None, ret='all')

returns a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict or scalar

static get_class_prefix()

return the token type prefix

static get_class_type()

return the class type identifier

static realms_dict_to_string(realms)

This function converts the realms - if it is a dictionary - to a string.

{“realm1”: {“selected”: True,

“count”: 1 },

“realm2”: {“selected”: True,

“count”: 2} -> realm1:1,realm2:2

Parameters:realms (dict) – the realms as they are passed from the WebUI Returns:realms Return type:basestring

update(param)

This method is called during the initialization process. :param param: parameters from the token init :type param: dict :return: None

Certificate Token

class privacyidea.lib.tokens.certificatetoken.CertificateTokenClass(aToken)

Token to implement an X509 certificate. The certificate can be enrolled by sending a CSR to the server or the keypair is created by the server. If the server creates the keypair, the user can download a PKCS12 file. The OTP PIN is used as passphrase for the PKCS12 file.

privacyIDEA is capable of working with different CA connectors.

Valid parameters are request or certificate, both PEM encoded. If you pass a request you also need to pass the ca that should be used to sign the request. Passing a certificate just uploads the certificate to a new token object.

A certificate token can be created by an administrative task with the token/init api like this:

Example Initialization Request:

POST /auth HTTP/1.1
Host: example.com
Accept: application/json

type=certificate
user=cornelius
realm=realm1
request=<PEM encoded request>
ca=<name of the ca connector>

Example Initialization Request, key generation on servers side

In this case the certificate is created on behalf of another user.

POST /auth HTTP/1.1
Host: example.com
Accept: application/json

type=certificate
user=cornelius
realm=realm1
generate=1
ca=<name of the ca connector>

Example response:

HTTP/1.1 200 OK
Content-Type: application/json

{
   "detail": {
     "certificate": "...PEM..."
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": true
   },
   "version": "privacyIDEA unknown"
 }

get_as_dict()

This returns the token data as a dictionary. It is used to display the token list at /token/list.

The certificate token can add the PKCS12 file if it exists

Returns:The token data as dict Return type:dict

static get_class_info(key=None, ret='all')

returns a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict or scalar

static get_class_prefix()

static get_class_type()

get_init_detail(params=None, user=None)

At the end of the initialization we return the certificate and the PKCS12 file, if the private key exists.

hKeyRequired = False

revoke()

This revokes the token. We need to determine the CA, which issues the certificate, contact the connector and revoke the certificate

Some token types may revoke a token without locking it.

set_pin(pin, encrypt=False)

set the PIN of a token. The PIN of the certificate token is stored encrypted. It is used as passphrase for the PKCS12 file.

Parameters:

• pin (basestring) – the pin to be set for the token
• encrypt (bool) – If set to True, the pin is stored encrypted and can be retrieved from the database again

update(param)

This method is called during the initialization process. :param param: parameters from the token init :type param: dict :return: None

using_pin = False

Daplug Token

class privacyidea.lib.tokens.daplugtoken.DaplugTokenClass(a_token)

daplug token class implementation

check_otp(anOtpVal, counter=None, window=None, options=None)

checkOtp - validate the token otp against a given otpvalue

Parameters:

• anOtpVal (string, format: efekeiebekeh) – the otpvalue to be verified
• counter (int) – the counter state, that should be verified
• window (int) – the counter +window, which should be checked
• options (dict) – the dict, which could contain token specific info

Returns:

the counter state or -1

Return type:

int

check_otp_exist(otp, window=10)

checks if the given OTP value is/are values of this very token. This is used to autoassign and to determine the serial number of a token.

Parameters:

• otp (string) – the to be verified otp value
• window (int) – the lookahead window for the counter

Returns:

counter or -1 if otp does not exist

Return type:

int

static get_class_info(key=None, ret='all')

returns a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict or string

static get_class_prefix()

static get_class_type()

get_multi_otp(count=0, epoch_start=0, epoch_end=0, curTime=None, timestamp=None)

get_otp(current_time=None)

resync(otp1, otp2, options=None)

resync the token based on two otp values - external method to do the resync of the token

Parameters:

• otp1 (string) – the first otp value
• otp2 (string) – the second otp value
• options (dict or None) – optional token specific parameters

Returns:

counter or -1 if otp does not exist

Return type:

int

split_pin_pass(passw, user=None, options=None)

Email Token

class privacyidea.lib.tokens.emailtoken.EmailTokenClass(aToken)

Implementation of the EMail Token Class, that sends OTP values via SMTP. (Similar to SMSTokenClass)

EMAIL_ADDRESS_KEY = 'email'

check_otp(anOtpVal, counter=None, window=None, options=None)

check the otpval of a token against a given counter and the window

Parameters:passw (string) – the to be verified passw/pin Returns:counter if found, -1 if not found Return type:int

create_challenge(transactionid=None, options=None)

create a challenge, which is submitted to the user

Parameters:

• transactionid – the id of this challenge
• options – the request context parameters / data

Returns:

tuple of (bool, message and data) bool, if submit was successful message is submitted to the user data is preserved in the challenge attributes - additional attributes, which are displayed in the

output

static get_class_info(key=None, ret='all')

returns all or a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

:rtype : s.o.

static get_class_prefix()

static get_class_type()

return the generic token class identifier

is_challenge_request(passw, user=None, options=None)

check, if the request would start a challenge

We need to define the function again, to get rid of the is_challenge_request-decorator of the HOTP-Token

Parameters:

• passw – password, which might be pin or pin+otp
• options – dictionary of additional request parameters

Returns:

returns true or false

classmethod test_config(params=None)

update(param, reset_failcount=True)

update - process initialization parameters

Parameters:param (dict) – dict of initialization parameters Returns:nothing

HOTP Token

class privacyidea.lib.tokens.hotptoken.HotpTokenClass(db_token)

hotp token class implementation

check_otp(anOtpVal, counter=None, window=None, options=None)

check if the given OTP value is valid for this token.

Parameters:

• anOtpVal (string) – the to be verified otpvalue
• counter (int) – the counter state, that should be verified
• window (int) – the counter +window, which should be checked
• options (dict) – the dict, which could contain token specific info

Returns:

the counter state or -1

Return type:

int

check_otp_exist(otp, window=10, symetric=False, inc_counter=True)

checks if the given OTP value is/are values of this very token. This is used to autoassign and to determine the serial number of a token.

Parameters:

• otp (string) – the to be verified otp value
• window (int) – the lookahead window for the counter

Returns:

counter or -1 if otp does not exist

Return type:

int

generate_symmetric_key(server_component, client_component, options=None)

Generate a composite key from a server and client component using a PBKDF2-based scheme.

Parameters:

• server_component (hex string) – The component usually generated by privacyIDEA
• client_component (hex string) – The component usually generated by the client (e.g. smartphone)
• options –

Returns:

the new generated key as hex string

static get_class_info(key=None, ret='all')

returns a subtree of the token definition Is used by lib.token.get_token_info

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict

static get_class_prefix()

Return the prefix, that is used as a prefix for the serial numbers. :return: oath

static get_class_type()

return the token type shortname

Returns:‘hotp’ Return type:string

classmethod get_default_settings(params, logged_in_user=None, policy_object=None, client_ip=None)

This method returns a dictionary with default settings for token enrollment. These default settings are defined in SCOPE.USER and are hotp_hashlib, hotp_otplen. If these are set, the user will only be able to enroll tokens with these values.

The returned dictionary is added to the parameters of the API call. :param params: The call parameters :type params: dict :param logged_in_user: The logged_in_user dictionary with “role”,

“username” and “realm”

Parameters:

• policy_object (PolicyClass) – The policy_object
• client_ip (basestring) – The client IP address

Returns:

default parameters

get_init_detail(params=None, user=None)

to complete the token initialization some additional details should be returned, which are displayed at the end of the token initialization. This is the e.g. the enrollment URL for a Google Authenticator.

get_multi_otp(count=0, epoch_start=0, epoch_end=0, curTime=None, timestamp=None, counter_index=False)

return a dictionary of multiple future OTP values of the HOTP/HMAC token

WARNING: the dict that is returned contains a sequence number as key.

This it NOT the otp counter!

Parameters:

• count (int) – how many otp values should be returned
• epoch_start – Not used in HOTP
• epoch_end – Not used in HOTP
• curTime – Not used in HOTP
• timestamp – not used in HOTP
• counter_index – whether the counter should be used as index

Returns:

tuple of status: boolean, error: text and the OTP dictionary

get_otp(current_time=None)

return the next otp value

Parameters:curTime – Not Used in HOTP Returns:next otp value and PIN if possible Return type:tuple

static get_sync_timeout()

get the token sync timeout value

Returns:timeout value in seconds Return type:int

hashlib

is_challenge_request(passw, user=None, options=None)

check, if the request would start a challenge

• default: if the passw contains only the pin, this request would

trigger a challenge

• in this place as well the policy for a token is checked

Parameters:

• passw – password, which might be pin or pin+otp
• options – dictionary of additional request parameters

Returns:

returns true or false

is_previous_otp(otp, window=10)

Check if the OTP values was previously used.

Parameters:

• otp –
• window –

Returns:

resync(otp1, otp2, options=None)

resync the token based on two otp values

Parameters:

• otp1 (string) – the first otp value
• otp2 (string) – the second otp value
• options (dict or None) – optional token specific parameters

Returns:

counter or -1 if otp does not exist

Return type:

int

update(param, reset_failcount=True)

process the initialization parameters

Do we really always need an otpkey? the otpKey is handled in the parent class :param param: dict of initialization parameters :type param: dict

Returns:nothing

mOTP Token

class privacyidea.lib.tokens.motptoken.MotpTokenClass(db_token)

check_otp(anOtpVal, counter=None, window=None, options=None)

validate the token otp against a given otpvalue

Parameters:

• anOtpVal (string) – the to be verified otpvalue
• counter (int) – the counter state, that should be verified
• window (int) – the counter +window, which should be checked
• options (dict) – the dict, which could contain token specific info

Returns:

the counter state or -1

Return type:

int

static get_class_info(key=None, ret='all')

returns a subtree of the token definition Is used by lib.token.get_token_info

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

:rtype : dict or string

static get_class_prefix()

static get_class_type()

get_init_detail(params=None, user=None)

to complete the token normalisation, the response of the initialization should be build by the token specific method, the getInitDetails

update(param, reset_failcount=True)

update - process initialization parameters

Parameters:param (dict) – dict of initialization parameters Returns:nothing

OCRA Token

The OCRA token is the base OCRA functionality. Usually it is created by importing a CSV or PSKC file.

This code is tested in tests/test_lib_tokens_tiqr.

Implementation

class privacyidea.lib.tokens.ocratoken.OcraTokenClass(db_token)

The OCRA Token Implementation

check_otp(otpval, counter=None, window=None, options=None)

This function is invoked by TokenClass.check_challenge_response and checks if the given password matches the expected response for the given challenge.

Parameters:

• otpval – the password (pin + otp)
• counter – ignored
• window – ignored
• options – dictionary that must contain “challenge”

Returns:

>=0 if the challenge matches, -1 otherwise

create_challenge(transactionid=None, options=None)

This method creates a challenge, which is submitted to the user. The submitted challenge will be preserved in the challenge database.

If no transaction id is given, the system will create a transaction id and return it, so that the response can refer to this transaction.

Parameters:

• transactionid – the id of this challenge
• options (dict) – the request context parameters / data

Returns:

tuple of (bool, message, transactionid, attributes)

Return type:

tuple

The return tuple builds up like this: bool if submit was successful; message which is displayed in the JSON response; additional attributes, which are displayed in the JSON response.

static get_class_info(key=None, ret='all')

returns a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict or scalar

static get_class_prefix()

Return the prefix, that is used as a prefix for the serial numbers. :return: OCRA :rtype: basestring

static get_class_type()

Returns the internal token type identifier :return: ocra :rtype: basestring

is_challenge_request(passw, user=None, options=None)

check, if the request would start a challenge In fact every Request that is not a response needs to start a challenge request.

At the moment we do not think of other ways to trigger a challenge.

This function is not decorated with

@challenge_response_allowed

as the OCRA token is always a challenge response token!

Parameters:

• passw – The PIN of the token.
• options – dictionary of additional request parameters

Returns:

returns true or false

update(param)

This method is called during the initialization process.

Parameters:param (dict) – parameters from the token init Returns:None

verify_response(passw=None, challenge=None)

This method verifies if the passw is the valid OCRA response to the challenge. In case of success we return a value > 0

Parameters:passw (string) – the password (pin+otp) Returns:return otp_counter. If -1, challenge does not match Return type:int

Paper Token

class privacyidea.lib.tokens.papertoken.PaperTokenClass(db_token)

The Paper Token allows to print out the next e.g. 100 OTP values. This sheet of paper can be used to authenticate and strike out the used OTP values.

static get_class_info(key=None, ret='all')

returns a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict or scalar

static get_class_prefix()

Return the prefix, that is used as a prefix for the serial numbers. :return: PPR

static get_class_type()

return the token type shortname

Returns:‘paper’ Return type:string

update(param, reset_failcount=True)

PasswordToken

class privacyidea.lib.tokens.passwordtoken.PasswordTokenClass(aToken)

This Token does use a fixed Password as the OTP value. In addition, the OTP PIN can be used with this token. This Token can be used for a scenario like losttoken

class SecretPassword(secObj)

check_password(password)

get_password()

PasswordTokenClass.check_otp(anOtpVal, counter=None, window=None, options=None)

This checks the static password

Parameters:anOtpVal – This contains the “OTP” value, which is the static

password :return: result of password check, 0 in case of success, -1 if fail :rtype: int

static PasswordTokenClass.get_class_info(key=None, ret='all')

returns a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict or scalar

static PasswordTokenClass.get_class_prefix()

static PasswordTokenClass.get_class_type()

PasswordTokenClass.set_otplen(otplen=0)

sets the OTP length to the length of the password

Parameters:otplen (int) – This is ignored in this class Result:None

PasswordTokenClass.update(param)

This method is called during the initialization process. :param param: parameters from the token init :type param: dict :return: None

Questionnaire Token

class privacyidea.lib.tokens.questionnairetoken.QuestionnaireTokenClass(db_token)

This is a Questionnaire Token. The token stores a list of questions and answers in the tokeninfo database table. The answers are encrypted. During authentication a random answer is selected and presented as challenge. The user has to remember and pass the right answer.

check_answer(given_answer, challenge_object)

Check if the given answer is the answer to the sent question. The question for this challenge response was stored in the challenge_object.

Then we get the answer from the tokeninfo.

Parameters:

• given_answer – The answer given by the user
• challenge_object – The challenge object as stored in the database

Returns:

in case of success: 1

check_challenge_response(user=None, passw=None, options=None)

This method verifies if there is a matching question for the given passw and also verifies if the answer is correct.

It then returns the the otp_counter = 1

Parameters:

• user (User object) – the requesting user
• passw (string) – the password - in fact it is the answer to the question
• options (dict) – additional arguments from the request, which could be token specific. Usually “transaction_id”

Returns:

return otp_counter. If -1, challenge does not match

Return type:

int

create_challenge(transactionid=None, options=None)

This method creates a challenge, which is submitted to the user. The submitted challenge will be preserved in the challenge database.

The challenge is a randomly selected question of the available questions for this token.

If no transaction id is given, the system will create a transaction id and return it, so that the response can refer to this transaction.

Parameters:

• transactionid – the id of this challenge
• options (dict) – the request context parameters / data

Returns:

tuple of (bool, message, transactionid, attributes)

Return type:

tuple

The return tuple builds up like this: bool if submit was successful; message which is displayed in the JSON response; additional attributes, which are displayed in the JSON response.

classmethod get_class_info(key=None, ret='all')

returns a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict or scalar

static get_class_prefix()

Return the prefix, that is used as a prefix for the serial numbers. :return: QUST :rtype: basestring

static get_class_type()

Returns the internal token type identifier :return: qust :rtype: basestring

static get_setting_type(key)

The setting type of questions is public, so that the user can also read the questions.

Parameters:key – The key of the setting Returns:“public” string

is_challenge_request(passw, user=None, options=None)

The questionnaire token is always a challenge response token. The challenge is triggered by providing the PIN as the password.

Parameters:

• passw (string) – password, which might be pin or pin+otp
• user (User object) – The user from the authentication request
• options (dict) – dictionary of additional request parameters

Returns:

true or false

Return type:

bool

update(param)

This method is called during the initialization process.

Parameters:param (dict) – parameters from the token init Returns:None

RADIUS Token

class privacyidea.lib.tokens.radiustoken.RadiusTokenClass(db_token)

check_otp(otpval, counter=None, window=None, options=None)

run the RADIUS request against the RADIUS server

Parameters:

• otpval – the OTP value
• counter (int) – The counter for counter based otp values
• window – a counter window
• options (dict) – additional token specific options

Returns:

counter of the matching OTP value.

Return type:

int

check_pin_local

lookup if pin should be checked locally or on radius host

Returns:bool

static get_class_info(key=None, ret='all')

returns a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict or string

static get_class_prefix()

static get_class_type()

split_pin_pass(passw, user=None, options=None)

Split the PIN and the OTP value. Only if it is locally checked and not remotely.

update(param)

Registration Code Token

class privacyidea.lib.tokens.registrationtoken.RegistrationTokenClass(aToken)

Token to implement a registration code. It can be used to create a registration code or a “TAN” which can be used once by a user to authenticate somewhere. After this registration code is used, the token is automatically deleted.

The idea is to provide a workflow, where the user can get a registration code by e.g. postal mail and then use this code as the initial first factor to authenticate to the UI to enroll real tokens.

A registration code can be created by an administrative task with the token/init api like this:

Example Authentication Request:

POST /token/init HTTP/1.1
Host: example.com
Accept: application/json

type=register
user=cornelius
realm=realm1

Example response:

HTTP/1.1 200 OK
Content-Type: application/json

{
   "detail": {
     "registrationcode": "12345808124095097608"
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": true
   },
   "version": "privacyIDEA unknown"
 }

static get_class_info(key=None, ret='all')

returns a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict or scalar

static get_class_prefix()

static get_class_type()

get_init_detail(params=None, user=None)

At the end of the initialization we return the registration code.

inc_count_auth_success()

Increase the counter, that counts successful authentications In case of successful authentication the token does needs to be deleted.

update(param)

This method is called during the initialization process. :param param: parameters from the token init :type param: dict :return: None

Remote Token

class privacyidea.lib.tokens.remotetoken.RemoteTokenClass(db_token)

The Remote token forwards an authentication request to another privacyIDEA server. The request can be forwarded to a user on the other server or to a serial number on the other server. The PIN can be checked on the local privacyIDEA server or on the remote server.

Using the Remote token you can assign one physical token to many different users.

authenticate(passw, user=None, options=None)

do the authentication on base of password / otp and user and options, the request parameters.

Here we contact the other privacyIDEA server to validate the OtpVal.

Parameters:

• passw – the password / otp
• user – the requesting user
• options – the additional request parameters

Returns:

tuple of (success, otp_count - 0 or -1, reply)

check_otp(otpval, counter=None, window=None, options=None)

run the http request against the remote host

Parameters:

• otpval – the OTP value
• counter (int) – The counter for counter based otp values
• window – a counter window
• options (dict) – additional token specific options

Returns:

counter of the matching OTP value.

Return type:

int

check_pin_local

lookup if pin should be checked locally or on remote host

Returns:bool

static get_class_info(key=None, ret='all')

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict or string

static get_class_prefix()

return the token type prefix

static get_class_type()

return the class type identifier

is_challenge_request(passw, user=None, options=None)

This method checks, if this is a request, that triggers a challenge. It depends on the way, the pin is checked - either locally or remote

Parameters:

• passw (string) – password, which might be pin or pin+otp
• user (User object) – The user from the authentication request
• options (dict) – dictionary of additional request parameters

Returns:

true or false

update(param)

second phase of the init process - updates parameters

Parameters:param – the request parameters Returns:

• nothing -

SMS Token

class privacyidea.lib.tokens.smstoken.SmsTokenClass(db_token)

The SMS token sends an SMS containing an OTP via some kind of gateway. The gateways can be an SMTP or HTTP gateway or the special sipgate protocol. The Gateways are defined in the SMSProvider Modules.

The SMS token is a challenge response token. I.e. the first request needs to contain the correct OTP PIN. If the OTP PIN is correct, the sending of the SMS is triggered. The second authentication must either contain the OTP PIN and the OTP value or the transaction_id and the OTP value.

Example 1st Authentication Request:

POST /validate/check HTTP/1.1
Host: example.com
Accept: application/json

user=cornelius
pass=otppin

Example 1st response:

HTTP/1.1 200 OK
Content-Type: application/json

{
   "detail": {
     "transaction_id": "xyz"
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": false
   },
   "version": "privacyIDEA unknown"
 }

After this, the SMS is triggered. When the SMS is received the second part of authentication looks like this:

Example 2nd Authentication Request:

POST /validate/check HTTP/1.1
Host: example.com
Accept: application/json

user=cornelius
transaction_id=xyz
pass=otppin

Example 1st response:

HTTP/1.1 200 OK
Content-Type: application/json

{
   "detail": {
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
     "status": true,
     "value": true
   },
   "version": "privacyIDEA unknown"
 }

check_otp(anOtpVal, counter=None, window=None, options=None)

check the otpval of a token against a given counter and the window

Parameters:passw (string) – the to be verified passw/pin Returns:counter if found, -1 if not found Return type:int

create_challenge(transactionid=None, options=None)

create a challenge, which is submitted to the user

Parameters:

• transactionid – the id of this challenge
• options – the request context parameters / data

Returns:

tuple of (bool, message and data) bool, if submit was successful message is submitted to the user data is preserved in the challenge attributes - additional attributes, which are displayed in the

output

static get_class_info(key=None, ret='all')

returns all or a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

:rtype : s.o.

static get_class_prefix()

static get_class_type()

return the generic token class identifier

is_challenge_request(passw, user=None, options=None)

check, if the request would start a challenge

We need to define the function again, to get rid of the is_challenge_request-decorator of the HOTP-Token

Parameters:

• passw – password, which might be pin or pin+otp
• options – dictionary of additional request parameters

Returns:

returns true or false

update(param, reset_failcount=True)

process initialization parameters

Parameters:param (dict) – dict of initialization parameters Returns:nothing

SPass Token

class privacyidea.lib.tokens.spasstoken.SpassTokenClass(db_token)

This is a simple pass token. It does have no OTP component. The OTP checking will always succeed. Of course, an OTP PIN can be used.

authenticate(passw, user=None, options=None)

in case of a wrong passw, we return a bad matching pin, so the result will be an invalid token

check_otp(otpval, counter=None, window=None, options=None)

As we have no otp value we always return true. (counter == 0)

static get_class_info(key=None, ret='all')

returns a subtree of the token definition Is used by lib.token.get_token_info

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict

static get_class_prefix()

static get_class_type()

static is_challenge_request(passw, user, options=None)

The spass token does not support challenge response :param passw: :param user: :param options: :return:

static is_challenge_response(passw, user, options=None, challenges=None)

update(param)

SSHKey Token

class privacyidea.lib.tokens.sshkeytoken.SSHkeyTokenClass(db_token)

The SSHKeyTokenClass provides a TokenClass that stores the public SSH key and can give the public SSH key via the getotp function. This can be used to manage SSH keys and retrieve the public ssh key to import it to authorized keys files.

static get_class_info(key=None, ret='all')

returns a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dictionary

static get_class_prefix()

static get_class_type()

get_sshkey()

returns the public SSH key

Returns:SSH pub key Return type:string

mode = ['authenticate']

update(param)

The key holds the public ssh key and this is required

The key probably is of the form “ssh-rsa BASE64 comment”

using_pin = False

TiQR Token

The TiQR token is a special App based token, which allows easy login and which is based on OCRA.

It generates an enrollment QR code, which contains a link with the more detailed enrollment information.

For a description of the TiQR protocol see

• https://www.usenix.org/legacy/events/lisa11/tech/full_papers/Rijswijk.pdf
• https://github.com/SURFnet/tiqr/wiki/Protocol-documentation.
• https://tiqr.org

The TiQR token is based on the OCRA algorithm. It lets you authenticate with your smartphone by scanning a QR code.

The TiQR token is enrolled via /token/init, but it requires no otpkey, since the otpkey is generated on the smartphone and pushed to the privacyIDEA server in a seconds step.

Enrollment

1. Start enrollment with /token/init
2. Scan the QR code in the details of the JSON result. The QR code contains a link to /ttype/tiqr?action=metadata
3. The TiQR Smartphone App will fetch this link and get more information
4. The TiQR Smartphone App will push the otpkey to a link /ttype/tiqr?action=enrollment and the token will be ready for use.

Authentication

An application that wants to use the TiQR token with privacyIDEA has to use the token in challenge response.

1. Call /validate/check?user=<user>&pass=<pin> with the PIN of the TiQR token
2. The details of the JSON response contain a QR code, that needs to be shown to the user. In addition the application needs to save the transaction_id in the response.
3. The user scans the QR code.
4. The TiQR App communicates with privacyIDEA via the API /ttype/tiqr. In this step the response of the App to the challenge is verified. The successful authentication is stored in the Challenge DB table. (No need for the application to take any action)
5. Now, the application needs to poll /validate/check?user=<user>&transaction_id=*&pass= to verifiy the successful authentication. The pass can be empty. If value=true is returned, the user authenticated successfully with the TiQR token.

This code is tested in tests/test_lib_tokens_tiqr.

Implementation

class privacyidea.lib.tokens.tiqrtoken.TiqrTokenClass(db_token)

The TiQR Token implementation.

static api_endpoint(request, g)

This provides a function to be plugged into the API endpoint /ttype/<tokentype> which is defined in api/ttype.py See Tokentype endpoints.

Parameters:

• request – The Flask request
• g – The Flask global object g

Returns:

Flask Response or text

check_challenge_response(user=None, passw=None, options=None)

This function checks, if the challenge for the given transaction_id was marked as answered correctly. For this we check the otp_status of the challenge with the transaction_id in the database.

We do not care about the password

Parameters:

• user (User object) – the requesting user
• passw (string) – the password (pin+otp)
• options (dict) – additional arguments from the request, which could be token specific. Usually “transaction_id”

Returns:

return otp_counter. If -1, challenge does not match

Return type:

int

create_challenge(transactionid=None, options=None)

This method creates a challenge, which is submitted to the user. The submitted challenge will be preserved in the challenge database.

If no transaction id is given, the system will create a transaction id and return it, so that the response can refer to this transaction.

Parameters:

• transactionid – the id of this challenge
• options (dict) – the request context parameters / data

Returns:

tuple of (bool, message, transactionid, attributes)

Return type:

tuple

The return tuple builds up like this: bool if submit was successful; message which is displayed in the JSON response; additional attributes, which are displayed in the JSON response.

static get_class_info(key=None, ret='all')

returns a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict or scalar

static get_class_prefix()

Return the prefix, that is used as a prefix for the serial numbers. :return: TiQR :rtype: basestring

static get_class_type()

Returns the internal token type identifier :return: tiqr :rtype: basestring

get_init_detail(params=None, user=None)

At the end of the initialization we return the URL for the TiQR App.

update(param)

This method is called during the initialization process.

Parameters:param (dict) – parameters from the token init Returns:None

TOTP Token

class privacyidea.lib.tokens.totptoken.TotpTokenClass(db_token)

check_otp(anOtpVal, counter=None, window=None, options=None)

validate the token otp against a given otpvalue

Parameters:

• anOtpVal (string) – the to be verified otpvalue
• counter – the counter state, that should be verified. For TOTP

this is the unix system time (seconds) divided by 30/60 :type counter: int :param window: the counter +window (sec), which should be checked :type window: int :param options: the dict, which could contain token specific info :type options: dict :return: the counter or -1 :rtype: int

check_otp_exist(otp, window=None, options=None, symetric=True, inc_counter=True)

checks if the given OTP value is/are values of this very token at all. This is used to autoassign and to determine the serial number of a token. In fact it is a check_otp with an enhanced window.

Parameters:

• otp (string) – the to be verified otp value
• window (int) – the lookahead window for the counter in seconds!!!

Returns:

counter or -1 if otp does not exist

Return type:

int

static get_class_info(key=None, ret='all')

returns a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict or scalar

static get_class_prefix()

Return the prefix, that is used as a prefix for the serial numbers. :return: TOTP

static get_class_type()

return the token type shortname

Returns:‘totp’ Return type:string

classmethod get_default_settings(params, logged_in_user=None, policy_object=None, client_ip=None)

This method returns a dictionary with default settings for token enrollment. These default settings are defined in SCOPE.USER and are totp_hashlib, totp_timestep and totp_otplen. If these are set, the user will only be able to enroll tokens with these values.

The returned dictionary is added to the parameters of the API call. :param params: The call parameters :type params: dict :param logged_in_user: The logged_in_user dictionary with “role”,

“username” and “realm”

Parameters:

• policy_object (PolicyClass) – The policy_object
• client_ip (basestring) – The client IP address

Returns:

default parameters

get_multi_otp(count=0, epoch_start=0, epoch_end=0, curTime=None, timestamp=None)

return a dictionary of multiple future OTP values of the HOTP/HMAC token

Parameters:

• count (int) – how many otp values should be returned
• epoch_start – not implemented
• epoch_end – not implemented
• curTime (datetime) – Simulate the servertime
• timestamp (epoch time) – Simulate the servertime

Returns:

tuple of status: boolean, error: text and the OTP dictionary

get_otp(current_time=None, do_truncation=True, time_seconds=None, challenge=None)

get the next OTP value

Parameters:current_time – the current time, for which the OTP value

should be calculated for. :type current_time: datetime object :param time_seconds: the current time, for which the OTP value should be calculated for (date +%s) :type: time_seconds: int, unix system time seconds :return: next otp value, and PIN, if possible :rtype: tuple

static get_setting_type(key)

hashlib

resync(otp1, otp2, options=None)

resync the token based on two otp values external method to do the resync of the token

Parameters:

• otp1 (string) – the first otp value
• otp2 (string) – the second otp value
• options (dict or None) – optional token specific parameters

Returns:

counter or -1 if otp does not exist

Return type:

int

resyncDiffLimit = 1

timeshift

timestep

timewindow

update(param, reset_failcount=True)

This is called during initialization of the token to add additional attributes to the token object.

Parameters:param (dict) – dict of initialization parameters Returns:nothing

U2F Token

U2F is the “Universal 2nd Factor” specified by the FIDO Alliance. The register and authentication process is described here:

https://fidoalliance.org/specs/fido-u2f-v1.0-nfc-bt-amendment-20150514/fido-u2f-raw-message-formats.html

But you do not need to be aware of this. privacyIDEA wraps all FIDO specific communication, which should make it easier for you, to integrate the U2F tokens managed by privacyIDEA into your application.

U2F Tokens can be either

• registered by administrators for users or
• registered by the users themselves.

Enrollment

The enrollment/registering can be completely performed within privacyIDEA.

But if you want to enroll the U2F token via the REST API you need to do it in two steps:

1. Step

POST /token/init HTTP/1.1
Host: example.com
Accept: application/json

type=utf

This step returns a serial number.

2. Step

POST /token/init HTTP/1.1
Host: example.com
Accept: application/json

type=utf
serial=U2F1234578
clientdata=<clientdata>
regdata=<regdata>

clientdata and regdata are the values returned by the U2F device.

You need to call the javascript function

u2f.register([registerRequest], [], function(u2fData) {} );

and the responseHandler needs to send the clientdata and regdata back to privacyIDEA (2. step).

Authentication

The U2F token is a challenge response token. I.e. you need to trigger a challenge e.g. by sending the OTP PIN/Password for this token.

Get the challenge

POST /validate/check HTTP/1.1
Host: example.com
Accept: application/json

user=cornelius
pass=tokenpin

Response

HTTP/1.1 200 OK
Content-Type: application/json

{
   "detail": {
     "attributes": {
                     "hideResponseInput": true,
                     "img": ...imageUrl...
                     "u2fSignRequest": {
                         "challenge": "...",
                         "appId": "...",
                         "keyHandle": "...",
                         "version": "U2F_V2"
                     }
                   },
     "message": "Please confirm with your U2F token (Yubico U2F EE ...)"
     "transaction_id": "02235076952647019161"
   },
   "id": 1,
   "jsonrpc": "2.0",
   "result": {
       "status": true,
       "value": false,
   },
   "version": "privacyIDEA unknown"
 }

Send the Response

The application now needs to call the javascript function u2f.sign with the u2fSignRequest from the response.

var signRequests = [ error.detail.attributes.u2fSignRequest ]; u2f.sign(signRequests, function(u2fResult) {} );

The response handler function needs to call the /validate/check API again with the signatureData and clientData returned by the U2F device in the u2fResult:

POST /validate/check HTTP/1.1
Host: example.com
Accept: application/json

user=cornelius
pass=
transaction_id=<transaction_id>
signaturedata=signatureData
clientdata=clientData

Implementation

class privacyidea.lib.tokens.u2ftoken.U2fTokenClass(db_token)

The U2F Token implementation.

static api_endpoint(request, g)

This provides a function to be plugged into the API endpoint /ttype/u2f

The u2f token can return the facet list at this URL.

Parameters:

• request – The Flask request
• g – The Flask global object g

Returns:

Flask Response or text

check_otp(otpval, counter=None, window=None, options=None)

This checks the response of a previous challenge. :param otpval: N/A :param counter: The authentication counter :param window: N/A :param options: contains “clientdata”, “signaturedata” and

“transaction_id”

Returns:A value > 0 in case of success

create_challenge(transactionid=None, options=None)

This method creates a challenge, which is submitted to the user. The submitted challenge will be preserved in the challenge database.

If no transaction id is given, the system will create a transaction id and return it, so that the response can refer to this transaction.

Parameters:

• transactionid – the id of this challenge
• options (dict) – the request context parameters / data

Returns:

tuple of (bool, message, transactionid, attributes)

Return type:

tuple

The return tuple builds up like this: bool if submit was successful; message which is displayed in the JSON response; additional attributes, which are displayed in the JSON response.

static get_class_info(key=None, ret='all')

returns a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict or scalar

static get_class_prefix()

Return the prefix, that is used as a prefix for the serial numbers. :return: U2F :rtype: basestring

static get_class_type()

Returns the internal token type identifier :return: u2f :rtype: basestring

get_init_detail(params=None, user=None)

At the end of the initialization we ask the user to press the button

is_challenge_request(passw, user=None, options=None)

check, if the request would start a challenge In fact every Request that is not a response needs to start a challenge request.

At the moment we do not think of other ways to trigger a challenge.

This function is not decorated with

@challenge_response_allowed

as the U2F token is always a challenge response token!

Parameters:

• passw – The PIN of the token.
• options – dictionary of additional request parameters

Returns:

returns true or false

update(param, reset_failcount=True)

This method is called during the initialization process.

Parameters:param (dict) – parameters from the token init Returns:None

Vasco Token

Yubico Token

class privacyidea.lib.tokens.yubicotoken.YubicoTokenClass(db_token)

check_otp(anOtpVal, counter=None, window=None, options=None)

Here we contact the Yubico Cloud server to validate the OtpVal.

static get_class_info(key=None, ret='all')

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

dict or string

static get_class_prefix()

static get_class_type()

update(param)

Yubikey Token

class privacyidea.lib.tokens.yubikeytoken.YubikeyTokenClass(db_token)

The Yubikey Token in the Yubico AES mode

classmethod api_endpoint(request, g)

This provides a function to be plugged into the API endpoint /ttype/yubikey which is defined in api/ttype.py

The endpoint /ttype/yubikey is used for the Yubico validate request according to https://developers.yubico.com/yubikey-val/Validation_Protocol_V2.0.html

Parameters:

• request – The Flask request
• g – The Flask global object g

Returns:

Flask Response or text

Required query parameters

Query id:The id of the client to identify the correct shared secret Query otp:The OTP from the yubikey in the yubikey mode Query nonce:16-40 bytes of random data

Optional parameters h, timestamp, sl, timeout are not supported at the moment.

check_otp(anOtpVal, counter=None, window=None, options=None)

validate the token otp against a given otpvalue

Parameters:

• anOtpVal (string) – the to be verified otpvalue
• counter (int) – the counter state. It is not used by the Yubikey because the current counter value is sent encrypted inside the OTP value
• window (int) – the counter +window, which is not used in the Yubikey because the current counter value is sent encrypted inside the OTP, allowing a simple comparison between the encrypted counter value and the stored counter value
• options (dict) – the dict, which could contain token specific info

Returns:

the counter state or an error code (< 0):

-1 if the OTP is old (counter < stored counter) -2 if the private_uid sent in the OTP is wrong (different from the one stored with the token) -3 if the CRC verification fails :rtype: int

check_otp_exist(otp, window=None)

checks if the given OTP value is/are values of this very token. This is used to autoassign and to determine the serial number of a token.

static check_yubikey_pass(passw)

if the Token has set a PIN the user must also enter the PIN for authentication!

This checks the output of a yubikey in AES mode without providing the serial number. The first 12 (of 44) or 16 of 48) characters are the tokenid, which is stored in the tokeninfo yubikey.tokenid or the prefix yubikey.prefix.

Parameters:passw (string) – The password that consist of the static yubikey prefix and the otp Returns:True/False and the User-Object of the token owner Return type:dict

static get_class_info(key=None, ret='all')

returns a subtree of the token definition

Parameters:

• key (string) – subsection identifier
• ret (user defined) – default return value, if nothing is found

Returns:

subsection if key exists or user defined

Return type:

s.o.

static get_class_prefix()

static get_class_type()

is_challenge_request(passw, user=None, options=None)

This method checks, if this is a request, that triggers a challenge.

Parameters:

• passw (string) – password, which might be pin or pin+otp
• user (User object) – The user from the authentication request
• options (dict) – dictionary of additional request parameters

Returns:

true or false

update(param, reset_failcount=True)

class privacyidea.lib.tokenclass.TokenClass(db_token)

add_init_details(key, value)

(was addInfo) Adds information to a volatile internal dict

add_tokeninfo(key, value, value_type=None)

Add a key and a value to the DB tokeninfo :param key: :param value: :return:

classmethod api_endpoint(request, g)

This provides a function to be plugged into the API endpoint /ttype/<tokentype> which is defined in api/ttype.py

The method should return

return “json”, {}

or

return “text”, “OK”

Parameters:

• request – The Flask request
• g – The Flask global object g

Returns:

Flask Response or text

authenticate(passw, user=None, options=None)

High level interface which covers the check_pin and check_otp This is the method that verifies single shot authentication like they are done with push button tokens.

It is a high level interface to support other tokens as well, which do not have a pin and otp separation - they could overwrite this method

If the authentication succeeds an OTP counter needs to be increased, i.e. the OTP value that was used for this authentication is invalidated!

Parameters:

• passw (string) – the password which could be pin+otp value
• user (User object) – The authenticating user
• options (dict) – dictionary of additional request parameters

Returns:

returns tuple of 1. true or false for the pin match, 2. the otpcounter (int) and the 3. reply (dict) that will be added as

additional information in the JSON response of /validate/check.

Return type:

tuple

static challenge_janitor()

Just clean up all challenges, for which the expiration has expired.

Returns:None

check_all(message_list)

Perform all checks on the token. Returns False if the token is either: * auth counter exceeded * not active * fail counter exceeded * validity period exceeded

This is used in the function token.check_token_list

Parameters:message_list – A list of messages Returns:False, if any of the checks fail

check_auth_counter()

This function checks the count_auth and the count_auth_success. If the count_auth is less than count_auth_max and count_auth_success is less than count_auth_success_max it returns True. Otherwise False.

Returns:success if the counter is less than max Return type:bool

check_challenge_response(user=None, passw=None, options=None)

This method verifies if there is a matching challenge for the given passw and also verifies if the response is correct.

It then returns the new otp_counter of the token.

In case of success the otp_counter will be >= 0.

Parameters:

• user (User object) – the requesting user
• passw (string) – the password (pin+otp)
• options (dict) – additional arguments from the request, which could be token specific. Usually “transactionid”

Returns:

return otp_counter. If -1, challenge does not match

Return type:

int

check_failcount()

Checks if the failcounter is exceeded. It returns True, if the failcounter is less than maxfail :return: True or False

check_last_auth_newer(last_auth)

Check if the last successful authentication with the token is newer than the specified time delta which is passed as 10h, 7d or 1y.

It returns True, if the last authentication with this token is newer* than the specified delta.

Parameters:last_auth (basestring) – 10h, 7d or 1y Returns:bool

check_otp(otpval, counter=None, window=None, options=None)

This checks the OTP value, AFTER the upper level did the checkPIN

In the base class we do not know, how to calculate the OTP value. So we return -1. In case of success, we should return >=0, the counter

Parameters:

• otpval – the OTP value
• counter (int) – The counter for counter based otp values
• window – a counter window
• options (dict) – additional token specific options

Returns:

counter of the matching OTP value.

Return type:

int

check_otp_exist(otp, window=None)

checks if the given OTP value is/are values of this very token. This is used to autoassign and to determine the serial number of a token.

Parameters:

• otp – the OTP value
• window (int) – The look ahead window

Returns:

True or a value > 0 in case of success

check_pin(pin, user=None, options=None)

Check the PIN of the given Password. Usually this is only dependent on the token itself, but the user object can cause certain policies.

Each token could implement its own PIN checking behaviour.

Parameters:

• pin (string) – the PIN (static password component), that is to be checked.
• user (User object) – for certain PIN policies (e.g. checking against the user store) this is the user, whose password would be checked. But at the moment we are checking against the userstore in the decorator “auth_otppin”.
• options – the optional request parameters

Returns:

If the PIN is correct, return True

Return type:

bool

check_validity_period()

This checks if the datetime.datetime.now() is within the validity period of the token.

Returns:success Return type:bool

create_challenge(transactionid=None, options=None)

This method creates a challenge, which is submitted to the user. The submitted challenge will be preserved in the challenge database.

If no transaction id is given, the system will create a transaction id and return it, so that the response can refer to this transaction.

Parameters:

• transactionid – the id of this challenge
• options (dict) – the request context parameters / data

Returns:

tuple of (bool, message, transactionid, attributes)

Return type:

tuple

The return tuple builds up like this: bool if submit was successful; message which is displayed in the JSON response; additional attributes, which are displayed in the JSON response.

static decode_otpkey(otpkey, otpkeyformat)

Decode the otp key which is given in a specific format.

Supported formats:

• hex, in which the otpkey is returned verbatim
• base32check, which is specified in decode_base32check

In case the OTP key is malformed or if the format is unknown, a ParameterError is raised.

Parameters:

• otpkey – OTP key passed by the user
• otpkeyformat – “hex” or “base32check”

Returns:

hex-encoded otpkey

del_tokeninfo(key=None)

delete_token()

delete the database token

enable(enable=True)

generate_symmetric_key(server_component, client_component, options=None)

This method generates a symmetric key, from a server component and a client component. This key generation could be based on HMAC, KDF or even Diffie-Hellman.

The basic key-generation is simply replacing the last n byte of the server component with bytes of the client component.

Parameters:

• server_component (hex string) – The component usually generated by privacyIDEA
• client_component – The component usually generated by the client (e.g. smartphone)
• options –

Returns:

the new generated key as hex string

get_QRimage_data(response_detail)

FIXME: Do we really use this?

get_as_dict()

This returns the token data as a dictionary. It is used to display the token list at /token/list.

Returns:The token data as dict Return type:dict

static get_class_info(key=None, ret='all')

static get_class_prefix()

static get_class_type()

get_count_auth()

Return the number of all authentication tries

get_count_auth_max()

Return the number of maximum allowed authentications

get_count_auth_success()

Return the number of successful authentications

get_count_auth_success_max()

Return the maximum allowed successful authentications

get_count_window()

classmethod get_default_settings(params, logged_in_user=None, policy_object=None, client_ip=None)

This method returns a dictionary with default settings for token enrollment. These default settings depend on the token type and the defined policies.

The returned dictionary is added to the parameters of the API call. :param params: The call parameters :type params: dict :param logged_in_user: The logged_in_user dictionary with “role”,

“username” and “realm”

Parameters:policy_object (PolicyClass) – The policy_object Returns:default parameters

get_failcount()

static get_hashlib(hLibStr)

Returns a hashlib function for a given string :param hLibStr: the hashlib :type hLibStr: string :return: the hashlib :rtype: function

get_init_detail(params=None, user=None)

to complete the token initialization, the response of the initialisation should be build by this token specific method. This method is called from api/token after the token is enrolled

get_init_detail returns additional information after an admin/init like the QR code of an HOTP/TOTP token. Can be anything else.

Parameters:

• params (dict) – The request params during token creation token/init
• user (User object) – the user, token owner

Returns:

additional descriptions

Return type:

dict

get_init_details()

return the status of the token rollout

Returns:return the status dict. Return type:dict

get_max_failcount()

get_multi_otp(count=0, epoch_start=0, epoch_end=0, curTime=None, timestamp=None)

This returns a dictionary of multiple future OTP values of a token.

Parameters:

• count – how many otp values should be returned
• epoch_start – time based tokens: start when
• epoch_end – time based tokens: stop when
• curTime (datetime object) – current time for TOTP token (for selftest)
• timestamp (int) – unix time, current time for TOTP token (for selftest)

Returns:

True/False, error text, OTP dictionary

Return type:

Tuple

get_otp(current_time='')

The default token does not support getting the otp value will return a tuple of four values a negative value is a failure.

Returns:something like: (1, pin, otpval, combined)

get_otp_count()

get_otp_count_window()

get_otplen()

get_pin_hash_seed()

get_realms()

Return a list of realms the token is assigned to :return: realms :rtype:l list

get_serial()

static get_setting_type(key)

This function returns the type of the token specific config/setting. This way a tokenclass can define settings, that can be “public” or a “password”. If this setting is written to the database, the type of the setting is set automatically in set_privacyidea_config

The key name needs to start with the token type.

Parameters:key – The token specific setting key Returns:A string like “public”

get_sync_window()

get_tokeninfo(key=None, default=None)

return the complete token info or a single key of the tokeninfo. When returning the complete token info dictionary encrypted entries are not decrypted. If you want to receive a decrypted value, you need to call it directly with the key.

Parameters:

• key (string) – the key to return
• default (string) – the default value, if the key does not exist

Returns:

the value for the key

Return type:

int or string

get_tokentype()

get_type()

get_user_displayname()

Returns a tuple of a user identifier like user@realm and the displayname of “givenname surname”.

Returns:tuple

get_user_id()

get_validity_period_end()

returns the end of validity period (if set) if not set, “” is returned. :return: the end of the validity period :rtype: string

get_validity_period_start()

returns the start of validity period (if set) if not set, “” is returned. :return: the start of the validity period :rtype: string

hKeyRequired = False

inc_count_auth()

Increase the counter, that counts authentications - successful and unsuccessful

inc_count_auth_success()

Increase the counter, that counts successful authentications Also increase the auth counter

inc_failcount()

inc_otp_counter(counter=None, increment=1, reset=True)

Increase the otp counter and store the token in the database

Before increasing the token.count the token.count can be set using the parameter counter.

Parameters:

• counter (int) – if given, the token counter is first set to counter and then increased by increment
• increment (int) – increase the counter by this amount
• reset (bool) – reset the failcounter if set to True

Returns:

the new counter value

is_active()

is_challenge_request(passw, user=None, options=None)

This method checks, if this is a request, that triggers a challenge.

The default behaviour to trigger a challenge is, if the passw parameter only contains the correct token pin and the request contains a data or a challenge key i.e. if the options parameter contains a key data or challenge.

Each token type can decide on its own under which condition a challenge is triggered by overwriting this method.

please note: in case of pin policy == 2 (no pin is required) the check_pin would always return true! Thus each request containing a data or challenge would trigger a challenge!

The Challenge workflow is like this.

When an authentication request is issued, first it is checked if this is a request which will create a new challenge (is_challenge_request) or if this is a response to an existing challenge (is_challenge_response). In these two cases during request processing the following functions are called.

is_challenge_request or is_challenge_response

|

V V

create_challenge check_challenge

|

V V

challenge_janitor challenge_janitor

Parameters:

• passw (string) – password, which might be pin or pin+otp
• user (User object) – The user from the authentication request
• options (dict) – dictionary of additional request parameters

Returns:

true or false

Return type:

bool

is_challenge_response(passw, user=None, options=None)

This method checks, if this is a request, that is the response to a previously sent challenge.

The default behaviour to check if this is the response to a previous challenge is simply by checking if the request contains a parameter state or transactionid i.e. checking if the options parameter contains a key state or transactionid.

This method does not try to verify the response itself! It only determines, if this is a response for a challenge or not. The response is verified in check_challenge_response.

Parameters:

• passw (string) – password, which might be pin or pin+otp
• user (User object) – the requesting user
• options (dict) – dictionary of additional request parameters

Returns:

true or false

Return type:

bool

is_locked()

Check if the token is in a locked state A locked token can not be modified

Returns:True, if the token is locked.

is_orphaned()

Return True is the token is orphaned.

An orphaned token means, that it has a user assigned, but the user does not exist in the user store (anymore) :return: True / False

is_pin_change(password=False)

Returns true if the pin of the token needs to be changed. :param password: Whether the password needs to be changed. :type password: bool

Returns:True or False

is_previous_otp(otp, window=10)

checks if a given OTP value is a previous OTP value, that lies in the past or has a lower counter.

This is used in case of a failed authentication to return the information, that this OTP values was used previously and is invalid.

Parameters:

• otp (basestring) – The OTP value.
• window (int) – A counter window, how far we should look into the past.

Returns:

bool

is_revoked()

Check if the token is in the revoked state

Returns:True, if the token is revoked

mode = ['authenticate', 'challenge']

reset()

Reset the failcounter

resync(otp1, otp2, options=None)

revoke()

This revokes the token. By default it 1. sets the revoked-field 2. set the locked field 3. disables the token.

Some token types may revoke a token without locking it.

save()

Save the database token

set_count_auth(count)

Sets the counter for the occurred login attepms as key “count_auth” in token info :param count: a number :type count: int

set_count_auth_max(count)

Sets the counter for the maximum allowed login attempts as key “count_auth_max” in token info :param count: a number :type count: int

set_count_auth_success(count)

Sets the counter for the occurred successful logins as key “count_auth_success” in token info :param count: a number :type count: int

set_count_auth_success_max(count)

Sets the counter for the maximum allowed successful logins as key “count_auth_success_max” in token info :param count: a number :type count: int

set_count_window(countWindow)

set_defaults()

Set the default values on the database level

set_description(description)

Set the description on the database level

Parameters:description (string) – description of the token

set_failcount(failcount)

Set the failcounter in the database

set_hashlib(hashlib)

set_init_details(details)

set_maxfail(maxFail)

set_next_pin_change(diff=None, password=False)

Sets the timestamp for the next_pin_change. Provide a difference like 90d (90 days).

Either provider the :param diff: The time delta. :type diff: basestring :param password: Do no set next_pin_change but next_password_change :return: None

set_otp_count(otpCount)

set_otpkey(otpKey)

set_otplen(otplen)

set_pin(pin, encrypt=False)

set the PIN of a token. Usually the pin is stored in a hashed way. :param pin: the pin to be set for the token :type pin: basestring :param encrypt: If set to True, the pin is stored encrypted and

can be retrieved from the database again

set_pin_hash_seed(pinhash, seed)

set_realms(realms, add=False)

Set the list of the realms of a token. :param realms: realms the token should be assigned to :type realms: list :param add: if the realms should be added and not replaced :type add: boolean

set_so_pin(soPin)

set_sync_window(syncWindow)

set_tokeninfo(info)

Set the tokeninfo field in the DB. Old values will be deleted. :param info: dictionary with key and value :type info: dict :return:

set_type(tokentype)

Set the tokentype in this object and also in the underlying database-Token-object.

Parameters:tokentype (string) – The type of the token like HOTP or TOTP

set_user(user, report=None)

Set the user attributes (uid, resolvername, resolvertype) of a token.

Parameters:

• user – a User() object, consisting of loginname and realm
• report – tbdf.

Returns:

None

set_user_identifiers(uid, resolvername, resolvertype)

(was setUid) Set the user attributes of a token :param uid: The user id in the user source :param resolvername: The name of the resolver :param resolvertype: The type of the resolver :return: None

set_user_pin(userPin)

set_validity_period_end(end_date)

sets the end date of the validity period for a token :param end_date: the end date in the format YYYY-MM-DDTHH:MM+OOOO

if the format is wrong, the method will throw an exception

set_validity_period_start(start_date)

sets the start date of the validity period for a token :param start_date: the start date in the format YYYY-MM-DDTHH:MM+OOOO

if the format is wrong, the method will throw an exception

split_pin_pass(passw, user=None, options=None)

Split the password into the token PIN and the OTP value

take the given password and split it into the PIN and the OTP value. The splitting can be dependent of certain policies. The policies may depend on the user.

Each token type may define its own way to slit the PIN and the OTP value.

Parameters:

• passw – the password to split
• user (User object) – The user/owner of the token
• options (dict) – can be used be the token types.

Returns:

tuple of pin and otp value

Returns:

tuple of (split status, pin, otp value)

Return type:

tuple

status_validation_fail()

callback to enable a status change, if auth failed

status_validation_success()

callback to enable a status change, if auth succeeds

static test_config(params=None)

This method is used to test the token config. Some tokens require some special token configuration like the SMS-Token or the Email-Token. To test this configuration, this classmethod is used.

It takes token specific parameters and returns a tuple of a boolean and a result description.

Parameters:params (dict) – token specific parameters Returns:success, description Return type:tuple

update(param, reset_failcount=True)

Update the token object

Parameters:param – a dictionary with different params like keysize, description, genkey, otpkey, pin Type:param: dict

user

return the user (owner) of a token If the token has no owner assigned, we return None

Returns:The owner of the token Return type:User object

using_pin = True

Token Functions

This module contains all top level token functions. It depends on the models, lib.user and lib.tokenclass (which depends on the tokenclass implementations like lib.tokens.hotptoken)

This is the middleware/glue between the HTTP API and the database

privacyidea.lib.token.add_tokeninfo(serial, info, value=None, value_type=None, user=None)

Sets a token info field in the database. The info is a dict for each token of key/value pairs.

Parameters:

• serial (basestring) – The serial number of the token
• info – The key of the info in the dict
• value – The value of the info
• value_type – The type of the value. If set to “password” the value

is stored encrypted :type value_type: basestring :param user: The owner of the tokens, that should be modified :type user: User object :return: the number of modified tokens :rtype: int

privacyidea.lib.token.assign_token(serial, user, pin=None, encrypt_pin=False)

Assign token to a user. If the PIN is given, the PIN is reset.

Parameters:

• serial (basestring) – The serial number of the token
• user (User object) – The user, to whom the token should be assigned.
• pin (basestring) – The PIN for the newly assigned token.
• encrypt_pin (bool) – Whether the PIN should be stored in an encrypted way

Returns:

True if the token was assigned, in case of an error an exception

is thrown :rtype: bool

privacyidea.lib.token.check_otp(serial, otpval)

This function checks the OTP for a given serial number :param serial: :param otpval: :return:

privacyidea.lib.token.check_realm_pass(realm, passw, options=None)

This function checks, if the given passw matches any token in the given realm. This can be used for the 4-eyes token. Only tokens that are assigned are tested.

It returns the res True/False and a reply_dict, which contains the serial number of the matching token.

Parameters:

• realm – The realm of the user
• passw – The password containing PIN+OTP
• options (dict) – Additional options that are passed to the tokens

Returns:

tuple of bool and dict

privacyidea.lib.token.check_serial(serial)

This checks, if the given serial number can be used for a new token. it returns a tuple (result, new_serial) result being True if the serial does not exist, yet. new_serial is a suggestion for a new serial number, that does not exist, yet.

Parameters:serial – Seral number that is to be checked, if it can be used for

a new token. :type serial: string :result: bool and serial number :rtype: tuple

privacyidea.lib.token.check_serial_pass(serial, passw, options=None)

This function checks the otp for a given serial

If the OTP matches, True is returned and the otp counter is increased.

The function tries to determine the user (token owner), to derive possible additional policies from the user.

Parameters:

• serial (basestring) – The serial number of the token
• passw (basestring) – The password usually consisting of pin + otp
• options (dict) – Additional options. Token specific.

Returns:

tuple of result (True, False) and additional dict

Return type:

tuple

privacyidea.lib.token.check_token_list(tokenobject_list, passw, user=None, options=None)

this takes a list of token objects and tries to find the matching token for the given passw. It also tests, * if the token is active or * the max fail count is reached, * if the validity period is ok...

This function is called by check_serial_pass, check_user_pass and check_yubikey_pass.

Parameters:

• tokenobject_list – list of identified tokens
• passw – the provided passw (mostly pin+otp)
• user – the identified use - as class object
• options – additional parameters, which are passed to the token

Returns:

tuple of success and optional response

Return type:

(bool, dict)

privacyidea.lib.token.check_user_pass(user, passw, options=None)

This function checks the otp for a given user. It is called by the API /validate/check

If the OTP matches, True is returned and the otp counter is increased.

Parameters:

• user (User object) – The user who is trying to authenticate
• passw (basestring) – The password usually consisting of pin + otp
• options (dict) – Additional options. Token specific.

Returns:

tuple of result (True, False) and additional dict

Return type:

tuple

privacyidea.lib.token.copy_token_pin(serial_from, serial_to)

This function copies the token PIN from one token to the other token. This can be used for workflows like lost token.

In fact the PinHash and the PinSeed are transferred

Parameters:

• serial_from (basestring) – The token to copy from
• serial_to (basestring) – The token to copy to

Returns:

True. In case of an error raise an exception

Return type:

bool

privacyidea.lib.token.copy_token_realms(serial_from, serial_to)

Copy the realms of one token to the other token

Parameters:

• serial_from – The token to copy from
• serial_to – The token to copy to

Returns:

None

privacyidea.lib.token.copy_token_user(serial_from, serial_to)

This function copies the user from one token to the other token. In fact the user_id, resolver and resolver type are transferred.

Parameters:

• serial_from (basestring) – The token to copy from
• serial_to (basestring) – The token to copy to

Returns:

True. In case of an error raise an exception

Return type:

bool

privacyidea.lib.token.create_tokenclass_object(db_token)

(was createTokenClassObject) create a token class object from a given type If a tokenclass for this type does not exist, the function returns None.

Parameters:db_token (database token object) – the database referenced token Returns:instance of the token class object Return type:tokenclass object

privacyidea.lib.token.delete_tokeninfo(serial, key, user=None)

Delete a specific token info field in the database.

Parameters:

• serial (basestring) – The serial number of the token
• key – The key of the info in the dict
• value – The value of the info
• user (User object) – The owner of the tokens, that should be modified

Returns:

the number of tokens matching the serial and user. This number also includes tokens that did not have

the token info key set in the first place! :rtype: int

privacyidea.lib.token.enable_token(serial, enable=True, user=None)

Enable or disable a token. This can be checked with is_token_active

Enabling an already active token will return 0.

Parameters:

• serial (basestring) – The serial number of the token
• enable (bool) – False is the token should be disabled
• user (User object) – all tokens of the user will be enabled or disabled

Returns:

Number of tokens that were enabled/disabled

Return type:

privacyidea.lib.token.gen_serial(tokentype=None, prefix=None)

generate a serial for a given tokentype

Parameters:

• tokentype – the token type prefix is done by a lookup on the tokens
• prefix – A prefix to the serial number

Returns:

serial number

Return type:

string

privacyidea.lib.token.get_all_token_users()

return a dictionary with all tokens, that are assigned to users. This returns a dictionary with the key being the serial number of the token and the user information as dict.

Returns:dictionary of serial numbers Return type:dict

privacyidea.lib.token.get_dynamic_policy_definitions(scope=None)

This returns the dynamic policy definitions that come with the new loaded token classes.

Parameters:scope – an optional scope parameter. Only return the policies of

this scope. :return: The policy definition for the token or only for the scope.

privacyidea.lib.token.get_multi_otp(serial, count=0, epoch_start=0, epoch_end=0, curTime=None, timestamp=None)

This function returns a list of OTP values for the given Token. Please note, that the tokentype needs to support this function.

Parameters:

• serial (basestring) – the serial number of the token
• count – number of the next otp values (to be used with event or time based tokens)
• epoch_start – unix time start date (used with time based tokens)
• epoch_end – unix time end date (used with time based tokens)
• curTime (datetime) – Simulate the servertime
• timestamp (int) – Simulate the servertime (unix time in seconds)

Returns:

dictionary of otp values

Return type:

dictionary

privacyidea.lib.token.get_num_tokens_in_realm(realm, active=True)

This returns the number of tokens in one realm. :param realm: The name of the realm :type realm: basestring :param active: If only active tokens should be taken into account :type active: bool :return: The number of tokens in the realm :rtype: int

privacyidea.lib.token.get_otp(serial, current_time=None)

This function returns the current OTP value for a given Token. The tokentype needs to support this function. if the token does not support getting the OTP value, a -2 is returned.

Parameters:

• serial – serial number of the token
• current_time (datetime) – a fake servertime for testing of TOTP token

Returns:

tuple with (result, pin, otpval, passw)

Return type:

tuple

privacyidea.lib.token.get_realms_of_token(serial, only_first_realm=False)

This function returns a list of the realms of a token

Parameters:

• serial (basestring) – the serial number of the token
• only_first_realm (bool) – Wheather we should only return the first realm

Returns:

list of the realm names

Return type:

list

privacyidea.lib.token.get_serial_by_otp(token_list, otp='', window=10)

Returns the serial for a given OTP value The tokenobject_list would be created by get_tokens()

Parameters:

• token_list (list of token objects) – the list of token objects to be investigated
• otp – the otp value, that needs to be found
• window (int) – the window of search

Returns:

the serial for a given OTP value and the user

Return type:

basestring

privacyidea.lib.token.get_token_by_otp(token_list, otp='', window=10)

search the token in the token_list, that creates the given OTP value. The tokenobject_list would be created by get_tokens()

Parameters:

• token_list (list of token objects) – the list of token objects to be investigated
• otp (basestring) – the otp value, that needs to be found
• window (int) – the window of search

Returns:

The token, that creates this OTP value

Return type:

Tokenobject

privacyidea.lib.token.get_token_owner(serial)

returns the user object, to which the token is assigned. the token is identified and retrieved by it’s serial number

If the token has no owner, None is returned

In case the serial number matches several tokens (like when containing a wildcard), also None is returned.

Parameters:serial (basestring) – serial number of the token Returns:The owner of the token Return type:User object or None

privacyidea.lib.token.get_token_type(serial)

Returns the tokentype of a given serial number

Parameters:serial (string) – the serial number of the to be searched token Returns:tokentype Return type:string

privacyidea.lib.token.get_tokenclass_info(tokentype, section=None)

return the config definition of a dynamic token

Parameters:

• tokentype (basestring) – the tokentype of the token like “totp” or “hotp”
• section (basestring) – subsection of the token definition - optional

Returns:

dict - if nothing found an empty dict

Return type:

dict

privacyidea.lib.token.get_tokens(tokentype=None, realm=None, assigned=None, user=None, serial=None, active=None, resolver=None, rollout_state=None, count=False, revoked=None, locked=None, tokeninfo=None, maxfail=None)

(was getTokensOfType) This function returns a list of token objects of a * given type, * of a realm * or tokens with assignment or not * for a certain serial number or * for a User

E.g. thus you can get all assigned tokens of type totp.

Parameters:

• tokentype (basestring) – The type of the token. If None, all tokens are returned.
• realm (basestring) – get tokens of a realm. If None, all tokens are returned.
• assigned (bool) – Get either assigned (True) or unassigned (False) tokens. If None get all tokens.
• user (User Object) – Filter for the Owner of the token
• serial (basestring) – The serial number of the token
• active (bool) – Whether only active (True) or inactive (False) tokens should be returned
• resolver (basestring) – filter for the given resolver name
• rollout_state – returns a list of the tokens in the certain rollout state. Some tokens are not enrolled in a single step but in multiple steps. These tokens are then identified by the DB-column rollout_state.
• count (bool) – If set to True, only the number of the result and not the list is returned.
• revoked (bool) – Only search for revoked tokens or only for not revoked tokens
• locked (bool) – Only search for locked tokens or only for not locked tokens
• tokeninfo (dict) – Return tokens with the given tokeninfo. The tokeninfo is a key/value dictionary
• maxfail – If only tokens should be returned, which failcounter reached maxfail

Returns:

A list of tokenclasses (lib.tokenclass)

Return type:

list

privacyidea.lib.token.get_tokens_in_resolver(resolver)

Return a list of the token ojects, that contain this very resolver

Parameters:resolver (basestring) – The resolver, the tokens should be in Returns:list of tokens with this resolver Return type:list of token objects

privacyidea.lib.token.get_tokens_paginate(tokentype=None, realm=None, assigned=None, user=None, serial=None, active=None, resolver=None, rollout_state=None, sortby=<sqlalchemy.orm.attributes.InstrumentedAttribute object>, sortdir='asc', psize=15, page=1, description=None, userid=None)

This function is used to retrieve a token list, that can be displayed in the Web UI. It supports pagination. Each retrieved page will also contain a “next” and a “prev”, indicating the next or previous page. If either does not exist, it is None.

Parameters:

• tokentype –
• realm –
• assigned (bool) – Returns assigned (True) or not assigned (False) tokens
• user (User object) – The user, whose token should be displayed
• serial –
• active –
• resolver (basestring) – A resolver name, which may contain “*” for filtering.
• userid (basestring) – A userid, which may contain “*” for filtering.
• rollout_state –
• sortby (A Token column or a string.) – Sort by a certain Token DB field. The default is Token.serial. If a string like “serial” is provided, we try to convert it to the DB column.
• sortdir (basestring) – Can be “asc” (default) or “desc”
• psize (int) – The size of the page
• page (int) – The number of the page to view. Starts with 1 ;-)

Returns:

dict with tokens, prev, next and count

Return type:

dict

privacyidea.lib.token.import_token(serial, token_dict, default_hashlib=None, tokenrealms=None)

This function is used during the import of a PSKC file.

Parameters:

• serial – The serial number of the token
• token_dict –

  A dictionary describing the token like: {“type”: ...,

  “description”: ..., “otpkey”: ..., “counter: ..., “timeShift”: ...}

• default_hashlib –

Returns:

the token object

privacyidea.lib.token.init_token(param, user=None, tokenrealms=None, tokenkind=None)

create a new token or update an existing token

Parameters:

• param (dict) – initialization parameters like: serial (optional) type (optionl, default=hotp) otpkey
• user (User Object) – the token owner
• tokenrealms (list) – the realms, to which the token should belong
• tokenkind – The kind of the token, can be “software”, “hardware” or “virtual”

Returns:

token object or None

Return type:

TokenClass object

privacyidea.lib.token.is_token_active(serial)

Return True if the token is active, otherwise false Returns None, if the token does not exist.

Parameters:serial (basestring) – The serial number of the token Returns:True or False Return type:bool

privacyidea.lib.token.is_token_owner(serial, user)

Check if the given user is the owner of the token with the given serial number :param serial: The serial number of the token :type serial: str :param user: The user that needs to be checked :type user: User object :return: Return True or False :rtype: bool

privacyidea.lib.token.lost_token(serial, new_serial=None, password=None, validity=10, contents='Ccns', pw_len=16, options=None)

This is the workflow to handle a lost token. The token <serial> is lost and will be disabled. A new token of type password token will be created and assigned to the user. The PIN of the lost token will be copied to the new token. The new token will have a certain validity period.

Parameters:

• serial – Token serial number
• new_serial – new serial number
• password – new password
• validity (int) – Number of days, the new token should be valid
• contents – The contents of the generated password. “C”: upper case

characters, “c”: lower case characters, “n”: digits and “s”: special characters :type contents: A string like “Ccn” :param pw_len: The length of the generated password :type pw_len: int :param options: optional values for the decorator passed from the upper API level :type options: dict

Returns:result dictionary

privacyidea.lib.token.remove_token(serial=None, user=None)

remove the token that matches the serial number or all tokens of the given user and also remove the realm associations and all its challenges

Parameters:

• user (User object) – The user, who’s tokens should be deleted.
• serial (basestring) – The serial number of the token to delete

Returns:

The number of deleted token

Return type:

int

privacyidea.lib.token.reset_token(serial, user=None)

Reset the failcounter :param serial: :param user: :return: The number of tokens, that were resetted :rtype: int

privacyidea.lib.token.resync_token(serial, otp1, otp2, options=None, user=None)

Resyncronize the token of the given serial number by searching the otp1 and otp2 in the future otp values.

Parameters:

• serial (basestring) – token serial number
• otp1 (basestring) – first OTP value
• otp2 (basestring) – second OTP value, directly after the first
• options (dict) – additional options like the servertime for TOTP token

Returns:

privacyidea.lib.token.revoke_token(serial, user=None)

Revoke a token.

Parameters:

• serial (basestring) – The serial number of the token
• enable (bool) – False is the token should be disabled
• user (User object) – all tokens of the user will be enabled or disabled

Returns:

Number of tokens that were enabled/disabled

Return type:

privacyidea.lib.token.set_count_auth(serial, count, user=None, max=False, success=False)

The auth counters are stored in the token info database field. There are different counters, that can be set

count_auth -> max=False, success=False count_auth_max -> max=True, success=False count_auth_success -> max=False, success=True count_auth_success_max -> max=True, success=True

Parameters:

• count (int) – The counter value
• user (User object) – The user owner of the tokens tokens to modify
• serial (basestring) – The serial number of the one token to modifiy
• max – True, if either count_auth_max or count_auth_success_max are

to be modified :type max: bool :param success: True, if either count_auth_success or count_auth_success_max are to be modified :type success: bool :return: number of modified tokens :rtype: int

privacyidea.lib.token.set_count_window(serial, countwindow=10, user=None)

The count window is used during authentication to find the matching OTP value. This sets the count window per token.

Parameters:

• serial (basestring) – The serial number of the token
• countwindow (int) – the size of the window
• user (User object) – The owner of the tokens, which should be modified

Returns:

number of modified tokens

Return type:

int

privacyidea.lib.token.set_defaults(serial)

Set the default values for the token with the given serial number :param serial: token serial :type serial: basestring :return: None

privacyidea.lib.token.set_description(serial, description, user=None)

Set the description of a token

Parameters:

• serial (basestring) – The serial number of the token
• description (int) – The description for the token
• user (User object) – The owner of the tokens, which should be modified

Returns:

number of modified tokens

Return type:

int

privacyidea.lib.token.set_failcounter(serial, counter, user=None)

Set the fail counter of a token.

Parameters:

• serial – The serial number of the token
• counter – THe counter to which the fail counter should be set
• user – An optional user

Returns:

Number of tokens, where the fail counter was set.

privacyidea.lib.token.set_hashlib(serial, hashlib='sha1', user=None)

Set the hashlib in the tokeninfo. Can be something like sha1, sha256...

Parameters:

• serial (basestring) – The serial number of the token
• hashlib (basestring) – The hashlib of the token
• user (User object) – The User, for who’s token the hashlib should be set

Returns:

the number of token infos set

Return type:

int

privacyidea.lib.token.set_max_failcount(serial, maxfail, user=None)

Set the maximum fail counts of tokens. This is the maximum number a failed authentication is allowed.

Parameters:

• serial (basestring) – The serial number of the token
• maxfail (int) – The maximum allowed failed authentications
• user (User object) – The owner of the tokens, which should be modified

Returns:

number of modified tokens

Return type:

int

privacyidea.lib.token.set_otplen(serial, otplen=6, user=None)

Set the otp length of the token defined by serial or for all tokens of the user. The OTP length is usually 6 or 8.

Parameters:

• serial (basestring) – The serial number of the token
• otplen (int) – The length of the OTP value
• user (User object) – The owner of the tokens

Returns:

number of modified tokens

Return type:

int

privacyidea.lib.token.set_pin(serial, pin, user=None, encrypt_pin=False)

Set the token PIN of the token. This is the static part that can be used to authenticate.

Parameters:

• pin (basestring) – The pin of the token
• user – If the user is specified, the pins for all tokens of this

user will be set :type used: User object :param serial: If the serial is specified, the PIN for this very token will be set. :return: The number of PINs set (usually 1) :rtype: int

privacyidea.lib.token.set_pin_so(serial, so_pin, user=None)

Set the SO PIN of a smartcard. The SO Pin can be used to reset the PIN of a smartcard. The SO PIN is stored in the database, so that it could be used for automatic processes for User PIN resetting.

Parameters:

• serial (basestring) – The serial number of the token
• so_pin – The Security Officer PIN

Returns:

The number of SO PINs set. (usually 1)

Return type:

int

privacyidea.lib.token.set_pin_user(serial, user_pin, user=None)

This sets the user pin of a token. This just stores the information of the user pin for (e.g. an eTokenNG, Smartcard) in the database

Parameters:

• serial (basestring) – The serial number of the token
• user_pin (basestring) – The user PIN

Returns:

The number of PINs set (usually 1)

Return type:

int

privacyidea.lib.token.set_realms(serial, realms=None, add=False)

Set all realms of a token. This sets the realms new. I.e. it does not add realms. So realms that are not contained in the list will not be assigned to the token anymore.

Thus, setting realms=[] clears all realms assignments.

Parameters:

• serial (basestring) – the serial number of the token
• realms (list) – A list of realm names
• add (bool) – if the realms should be added and not replaced

Returns:

the number of tokens, to which realms where added. As a serial

number should be unique, this is either 1 or 0. :rtype: int

privacyidea.lib.token.set_sync_window(serial, syncwindow=1000, user=None)

The sync window is the window that is used during resync of a token. Such many OTP values are calculated ahead, to find the matching otp value and counter.

Parameters:

• serial (basestring) – The serial number of the token
• syncwindow (int) – The size of the sync window
• user (User object) – The owner of the tokens, which should be modified

Returns:

number of modified tokens

Return type:

int

privacyidea.lib.token.set_validity_period_end(serial, user, end)

Set the validity period for the given token.

Parameters:

• serial –
• user –
• end (basestring) – Timestamp in the format DD/MM/YY HH:MM

privacyidea.lib.token.set_validity_period_start(serial, user, start)

Set the validity period for the given token.

Parameters:

• serial –
• user –
• start (basestring) – Timestamp in the format DD/MM/YY HH:MM

privacyidea.lib.token.token_exist(serial)

returns true if the token with the given serial number exists

Parameters:serial – the serial number of the token

privacyidea.lib.token.unassign_token(serial, user=None)

unassign the user from the token

Parameters:serial – The serial number of the token to unassign Returns:True

Application Class

privacyidea.lib.applications.MachineApplicationBase

alias of MachineApplication

Policy Module

Base function to handle the policy entries in the database. This module only depends on the db/models.py

The functions of this module are tested in tests/test_lib_policy.py

A policy has the attributes

• name
• scope
• action
• realm
• resolver
• user
• client
• active

name is the unique identifier of a policy. scope is the area, where this policy is meant for. This can be values like admin, selfservice, authentication... scope takes only one value.

active is bool and indicates, whether a policy is active or not.

action, realm, resolver, user and client can take a comma separated list of values.

realm and resolver

If these are empty ‘*’, this policy matches each requested realm.

user

If the user is empty or ‘*’, this policy matches each user. You can exclude users from matching this policy, by prepending a ‘-‘ or a ‘!’. *, -admin will match for all users except the admin.

You can also use regular expressions to match the user like customer_.* to match any user, starting with customer_.

Note

Regular expression will only work for exact machtes. user1234 will not match user1 but only user1...

client

The client is identified by its IP address. A policy can contain a list of IP addresses or subnets. You can exclude clients from subnets by prepending the client with a ‘-‘ or a ‘!’. 172.16.0.0/24, -172.16.0.17 will match each client in the subnet except the 172.16.0.17.

time

You can specify a time in which the policy should be active. Time formats are

<dow>-<dow>:<hh>:<mm>-<hh>:<mm>, ... <dow>:<hh>:<mm>-<hh>:<mm> <dow>:<hh>-<hh>

and any combination of it. “dow” being day of week Mon, Tue, Wed, Thu, Fri, Sat, Sun.

class privacyidea.lib.policy.ACTION

This is the list of usual actions.

ADDRESOLVERINRESPONSE = 'add_resolver_in_response'

ADDUSER = 'adduser'

ADDUSERINRESPONSE = 'add_user_in_response'

APIKEY = 'api_key_required'

ASSIGN = 'assign'

AUDIT = 'auditlog'

AUDIT_AGE = 'auditlog_age'

AUDIT_DOWNLOAD = 'auditlog_download'

AUTHITEMS = 'fetch_authentication_items'

AUTHMAXFAIL = 'auth_max_fail'

AUTHMAXSUCCESS = 'auth_max_success'

AUTH_CACHE = 'auth_cache'

AUTOASSIGN = 'autoassignment'

CACONNECTORDELETE = 'caconnectordelete'

CACONNECTORREAD = 'caconnectorread'

CACONNECTORWRITE = 'caconnectorwrite'

CHALLENGERESPONSE = 'challenge_response'

CHANGE_PIN_EVERY = 'change_pin_every'

CHANGE_PIN_FIRST_USE = 'change_pin_on_first_use'

CLIENTTYPE = 'clienttype'

CONFIGDOCUMENTATION = 'system_documentation'

COPYTOKENPIN = 'copytokenpin'

COPYTOKENUSER = 'copytokenuser'

CUSTOM_BASELINE = 'custom_baseline'

CUSTOM_MENU = 'custom_menu'

DEFAULT_TOKENTYPE = 'default_tokentype'

DELETE = 'delete'

DELETEUSER = 'deleteuser'

DISABLE = 'disable'

EMAILCONFIG = 'smtpconfig'

ENABLE = 'enable'

ENCRYPTPIN = 'encrypt_pin'

ENROLLPIN = 'enrollpin'

EVENTHANDLINGWRITE = 'eventhandling_write'

GETCHALLENGES = 'getchallenges'

GETRANDOM = 'getrandom'

GETSERIAL = 'getserial'

HIDE_WELCOME = 'hide_welcome_info'

IMPORT = 'importtokens'

LASTAUTH = 'last_auth'

LOGINMODE = 'login_mode'

LOGOUTTIME = 'logout_time'

LOSTTOKEN = 'losttoken'

LOSTTOKENPWCONTENTS = 'losttoken_PW_contents'

LOSTTOKENPWLEN = 'losttoken_PW_length'

LOSTTOKENVALID = 'losttoken_valid'

MACHINELIST = 'machinelist'

MACHINERESOLVERDELETE = 'mresolverdelete'

MACHINERESOLVERWRITE = 'mresolverwrite'

MACHINETOKENS = 'manage_machine_tokens'

MANAGESUBSCRIPTION = 'managesubscription'

MANGLE = 'mangle'

MAXTOKENREALM = 'max_token_per_realm'

MAXTOKENUSER = 'max_token_per_user'

NODETAILFAIL = 'no_detail_on_fail'

NODETAILSUCCESS = 'no_detail_on_success'

OTPPIN = 'otppin'

OTPPINCONTENTS = 'otp_pin_contents'

OTPPINMAXLEN = 'otp_pin_maxlength'

OTPPINMINLEN = 'otp_pin_minlength'

OTPPINRANDOM = 'otp_pin_random'

PASSNOTOKEN = 'passOnNoToken'

PASSNOUSER = 'passOnNoUser'

PASSTHRU = 'passthru'

PASSWORDRESET = 'password_reset'

PINHANDLING = 'pinhandling'

POLICYDELETE = 'policydelete'

POLICYTEMPLATEURL = 'policy_template_url'

POLICYWRITE = 'policywrite'

PRIVACYIDEASERVERWRITE = 'privacyideaserver_write'

RADIUSSERVERWRITE = 'radiusserver_write'

REALM = 'realm'

REALMDROPDOWN = 'realm_dropdown'

REGISTERBODY = 'registration_body'

REMOTE_USER = 'remote_user'

REQUIREDEMAIL = 'requiredemail'

RESET = 'reset'

RESETALLTOKENS = 'reset_all_user_tokens'

RESOLVER = 'resolver'

RESOLVERDELETE = 'resolverdelete'

RESOLVERWRITE = 'resolverwrite'

RESYNC = 'resync'

REVOKE = 'revoke'

SEARCH_ON_ENTER = 'search_on_enter'

SERIAL = 'serial'

SET = 'set'

SETHSM = 'set_hsm_password'

SETPIN = 'setpin'

SETREALM = 'setrealm'

SETTOKENINFO = 'settokeninfo'

SMSGATEWAYWRITE = 'smsgateway_write'

SMTPSERVERWRITE = 'smtpserver_write'

SYSTEMDELETE = 'configdelete'

SYSTEMWRITE = 'configwrite'

TIMEOUT_ACTION = 'timeout_action'

TOKENINFO = 'tokeninfo'

TOKENISSUER = 'tokenissuer'

TOKENLABEL = 'tokenlabel'

TOKENPAGESIZE = 'token_page_size'

TOKENREALMS = 'tokenrealms'

TOKENTYPE = 'tokentype'

TOKENWIZARD = 'tokenwizard'

TOKENWIZARD2ND = 'tokenwizard_2nd_token'

TRIGGERCHALLENGE = 'triggerchallenge'

UNASSIGN = 'unassign'

UPDATEUSER = 'updateuser'

USERDETAILS = 'user_details'

USERLIST = 'userlist'

USERPAGESIZE = 'user_page_size'

class privacyidea.lib.policy.ACTIONVALUE

This is a list of usual action values for e.g. policy action-values like otppin.

DISABLE = 'disable'

NONE = 'none'

TOKENPIN = 'tokenpin'

USERSTORE = 'userstore'

class privacyidea.lib.policy.AUTOASSIGNVALUE

This is the possible values for autoassign

NONE = 'any_pin'

USERSTORE = 'userstore'

class privacyidea.lib.policy.GROUP

These are the allowed policy action groups. The policies will be grouped in the UI.

ENROLLMENT = 'enrollment'

GENERAL = 'general'

MACHINE = 'machine'

PIN = 'pin'

SYSTEM = 'system'

TOKEN = 'token'

TOOLS = 'tools'

USER = 'user'

class privacyidea.lib.policy.LOGINMODE

This is the list of possible values for the login mode.

DISABLE = 'disable'

PRIVACYIDEA = 'privacyIDEA'

USERSTORE = 'userstore'

class privacyidea.lib.policy.MAIN_MENU

These are the allowed top level menu items. These are used to toggle the visibility of the menu items depending on the rights of the user

AUDIT = 'audit'

COMPONENTS = 'components'

CONFIG = 'config'

MACHINES = 'machines'

TOKENS = 'tokens'

USERS = 'users'

class privacyidea.lib.policy.PolicyClass

The Policy_Object will contain all database policy entries for easy filtering and mangling. It will be created at the beginning of the request and is supposed to stay alive unchanged during the request.

get_action_values(action, scope='authorization', realm=None, resolver=None, user=None, client=None, unique=False, allow_white_space_in_action=False, adminrealm=None)

Get the defined action values for a certain action like

scope: authorization action: tokentype

would return a list of the tokentypes

scope: authorization action: serial

would return a list of allowed serials

Parameters:

• unique – if set, the function will raise an exception if more than one value is returned
• allow_white_space_in_action (bool) – Some policies like emailtext would allow entering text with whitespaces. These whitespaces must not be used to separate action values!

Returns:

A list of the allowed tokentypes

Return type:

list

get_policies(name=None, scope=None, realm=None, active=None, resolver=None, user=None, client=None, action=None, adminrealm=None, time=None, all_times=False)

Return the policies of the given filter values

Parameters:

• name – The name of the policy
• scope – The scope of the policy
• realm – The realm in the policy
• active – Only active policies
• resolver – Only policies with this resolver
• user (basestring) – Only policies with this user
• client –
• action – Only policies, that contain this very action.
• adminrealm – This is the realm of the admin. This is only evaluated in the scope admin.
• time (datetime) – The optional time, for which the policies should be fetched. The default time is now()
• all_times (bool) – If True the time restriction of the policies is ignored. Policies of all time ranges will be returned.

Returns:

list of policies

Return type:

list of dicts

reload_from_db()

Read the timestamp from the database. If the timestamp is newer than the internal timestamp, then read the complete data :return:

ui_get_enroll_tokentypes(client, logged_in_user)

Return a dictionary of the allowed tokentypes for the logged in user. This used for the token enrollment UI.

It looks like this:

{“hotp”: “HOTP: event based One Time Passwords”,

“totp”: “TOTP: time based One Time Passwords”, “spass”: “SPass: Simple Pass token. Static passwords”, “motp”: “mOTP: classical mobile One Time Passwords”, “sshkey”: “SSH Public Key: The public SSH key”, “yubikey”: “Yubikey AES mode: One Time Passwords with Yubikey”, “remote”: “Remote Token: Forward authentication request to another server”, “yubico”: “Yubikey Cloud mode: Forward authentication request to YubiCloud”, “radius”: “RADIUS: Forward authentication request to a RADIUS server”, “email”: “EMail: Send a One Time Passwort to the users email address”, “sms”: “SMS: Send a One Time Password to the users mobile phone”, “certificate”: “Certificate: Enroll an x509 Certificate Token.”}

Parameters:

• client (basestring) – Client IP address
• logged_in_user (dict) – The Dict of the logged in user

Returns:

list of token types, the user may enroll

ui_get_main_menus(logged_in_user, client=None)

Get the list of allowed main menus derived from the policies for the given user - admin or normal user. It fetches all policies for this user and compiles a list of allowed menus to display or hide in the UI.

Parameters:

• logged_in_user – The logged in user, a dictionary with keys “username”, “realm” and “role”.
• client – The IP address of the client

Returns:

A list of MENUs to be displayed

ui_get_rights(scope, realm, username, client=None)

Get the rights derived from the policies for the given realm and user. Works for admins and normal users. It fetches all policies for this user and compiles a maximum list of allowed rights, that can be used to hide certain UI elements.

Parameters:

• scope – Can be SCOPE.ADMIN or SCOPE.USER
• realm – Is either user users realm or the adminrealm
• username – The loginname of the user
• client – The HTTP client IP

Returns:

A list of actions

class privacyidea.lib.policy.REMOTE_USER

The list of possible values for the remote_user policy.

ACTIVE = 'allowed'

DISABLE = 'disable'

class privacyidea.lib.policy.SCOPE

This is the list of the allowed scopes that can be used in policy definitions.

ADMIN = 'admin'

AUDIT = 'audit'

AUTH = 'authentication'

AUTHZ = 'authorization'

ENROLL = 'enrollment'

GETTOKEN = 'gettoken'

REGISTER = 'register'

USER = 'user'

WEBUI = 'webui'

class privacyidea.lib.policy.TIMEOUT_ACTION

This is a list of actions values for idle users

LOCKSCREEN = 'lockscreen'

LOGOUT = 'logout'

privacyidea.lib.policy.delete_all_policies()

privacyidea.lib.policy.delete_policy(name)

Function to delete one named policy

Parameters:name – the name of the policy to be deleted Returns:the count of the deleted policies. Return type:int

privacyidea.lib.policy.enable_policy(name, enable=True)

Enable or disable the policy with the given name :param name: :return: ID of the policy

privacyidea.lib.policy.export_policies(policies)

This function takes a policy list and creates an export file from it

Parameters:policies (list of policy dictionaries) – a policy definition Returns:the contents of the file Return type:string

privacyidea.lib.policy.get_static_policy_definitions(scope=None)

These are the static hard coded policy definitions. They can be enhanced by token based policy definitions, that can be found in lib.token.get_dynamic_policy_definitions.

Parameters:scope (basestring) – Optional the scope of the policies Returns:allowed scopes with allowed actions, the type of action and a

description. :rtype: dict

privacyidea.lib.policy.import_policies(file_contents)

This function imports policies from a file. The file has a config_object format, i.e. the text file has a header

[<policy_name>] key = value

and key value pairs.

Parameters:file_contents (basestring) – The contents of the file Returns:number of imported policies Return type:int

privacyidea.lib.policy.set_policy(name=None, scope=None, action=None, realm=None, resolver=None, user=None, time=None, client=None, active=True, adminrealm=None, check_all_resolvers=False)

Function to set a policy. If the policy with this name already exists, it updates the policy. It expects a dict of with the following keys: :param name: The name of the policy :param scope: The scope of the policy. Something like “admin”, “system”, “authentication” :param action: A scope specific action or a comma separated list of actions :type active: basestring :param realm: A realm, for which this policy is valid :param resolver: A resolver, for which this policy is valid :param user: A username or a list of usernames :param time: N/A if type() :param client: A client IP with optionally a subnet like 172.16.0.0/16 :param active: If the policy is active or not :type active: bool :param check_all_resolvers: If all the resolvers of a user should be

checked with this policy

Returns:The database ID od the the policy Return type:int

API Policies

Pre Policies

These are the policy decorators as PRE conditions for the API calls. I.e. these conditions are executed before the wrapped API call. This module uses the policy base functions from privacyidea.lib.policy but also components from flask like g.

Wrapping the functions in a decorator class enables easy modular testing.

The functions of this module are tested in tests/test_api_lib_policy.py

privacyidea.api.lib.prepolicy.allowed_audit_realm(request=None, action=None)

This decorator function takes the request and adds additional parameters to the request according to the policy for the SCOPE.ADMIN or ACTION.AUDIT :param request: :param action: :return: True

privacyidea.api.lib.prepolicy.api_key_required(request=None, action=None)

This is a decorator for check_user_pass and check_serial_pass. It checks, if a policy scope=auth, action=apikeyrequired is set. If so, the validate request will only performed, if a JWT token is passed with role=validate.

privacyidea.api.lib.prepolicy.auditlog_age(request=None, action=None)

This pre condition checks for the policy auditlog_age and set the “timelimit” parameter of the audit search API.

Check ACTION.AUDIT_AGE

The decorator can wrap GET /audit/

Parameters:

• request (Request Object) – The request that is intercepted during the API call
• action (basestring) – An optional Action

Returns:

Always true. Modified the parameter request

privacyidea.api.lib.prepolicy.check_anonymous_user(request=None, action=None)

This decorator function takes the request and verifies the given action for the SCOPE USER without an authenticated user but the user from the parameters.

This is used with password_reset

Parameters:

• request –
• action –

Returns:

True otherwise raises an Exception

privacyidea.api.lib.prepolicy.check_base_action(request=None, action=None, anonymous=False)

This decorator function takes the request and verifies the given action for the SCOPE ADMIN or USER. :param request: :param action: :param anonymous: If set to True, the user data is taken from the request

parameters.

Returns:True otherwise raises an Exception

privacyidea.api.lib.prepolicy.check_external(request=None, action='init')

This decorator is a hook to an external check function, that is called before the token/init or token/assign API.

Parameters:

• request (flask Request object) – The REST request
• action (basestring) – This is either “init” or “assign”

Returns:

either True or an Exception is raised

privacyidea.api.lib.prepolicy.check_max_token_realm(request=None, action=None)

Pre Policy This checks the maximum token per realm. Check ACTION.MAXTOKENREALM

This decorator can wrap:

/token/init (with a realm and user) /token/assign /token/tokenrealms

Parameters:

• req (Request Object) – The request that is intercepted during the API call
• action (basestring) – An optional Action

Returns:

True otherwise raises an Exception

privacyidea.api.lib.prepolicy.check_max_token_user(request=None, action=None)

Pre Policy This checks the maximum token per user policy. Check ACTION.MAXTOKENUSER

This decorator can wrap:

/token/init (with a realm and user) /token/assign

Parameters:

• req –
• action –

Returns:

True otherwise raises an Exception

privacyidea.api.lib.prepolicy.check_otp_pin(request=None, action=None)

This policy function checks if the OTP PIN that is about to be set follows the OTP PIN policies ACTION.OTPPINMAXLEN, ACTION.OTPPINMINLEN and ACTION.OTPPINCONTENTS and token-type-specific PIN policy actions in the SCOPE.USER or SCOPE.ADMIN. It is used to decorate the API functions.

The pin is investigated in the params as “otppin” or “pin”

In case the given OTP PIN does not match the requirements an exception is raised.

privacyidea.api.lib.prepolicy.check_token_init(request=None, action=None)

This decorator function takes the request and verifies if the requested tokentype is allowed to be enrolled in the SCOPE ADMIN or the SCOPE USER. :param request: :param action: :return: True or an Exception is raised

privacyidea.api.lib.prepolicy.check_token_upload(request=None, action=None)

This decorator function takes the request and verifies the given action for scope ADMIN :param req: :param filename: :return:

privacyidea.api.lib.prepolicy.encrypt_pin(request=None, action=None)

This policy function is to be used as a decorator for several API functions. E.g. token/assign, token/setpin, token/init If the policy is set to define the PIN to be encrypted, the request.all_data is modified like this: encryptpin = True

It uses the policy SCOPE.ENROLL, ACTION.ENCRYPTPIN

privacyidea.api.lib.prepolicy.enroll_pin(request=None, action=None)

This policy function is used as decorator for init token. It checks, if the user or the admin is allowed to set a token PIN during enrollment. If not, it deleted the PIN from the request.

privacyidea.api.lib.prepolicy.init_random_pin(request=None, action=None)

This policy function is to be used as a decorator in the API init function. If the policy is set accordingly it adds a random PIN to the request.all_data like.

It uses the policy SCOPE.ENROLL, ACTION.OTPPINRANDOM to set a random OTP PIN during Token enrollment

privacyidea.api.lib.prepolicy.init_token_defaults(request=None, action=None)

This policy function is used as a decorator for the API init function. Depending on policy settings it can add token specific default values like totp_hashlib, hotp_hashlib, totp_otplen...

privacyidea.api.lib.prepolicy.init_tokenlabel(request=None, action=None)

This policy function is to be used as a decorator in the API init function. It adds the tokenlabel definition to the params like this: params : { “tokenlabel”: “<u>@<r>” }

In addtion it adds the tokenissuer to the params like this: params : { “tokenissuer”: “privacyIDEA instance” }

It uses the policy SCOPE.ENROLL, ACTION.TOKENLABEL and ACTION.TOKENISSUER to set the tokenlabel and tokenissuer of Smartphone tokens during enrollment and this fill the details of the response.

privacyidea.api.lib.prepolicy.is_remote_user_allowed(req)

Checks if the REMOTE_USER server variable is allowed to be used.

Note

This is not used as a decorator!

Parameters:req – The flask request, containing the remote user and the client IP Returns:

privacyidea.api.lib.prepolicy.mangle(request=None, action=None)

This pre condition checks if either of the parameters pass, user or realm in a validate/check request should be rewritten based on an authentication policy with action “mangle”. See mangle for an example.

Check ACTION.MANGLE

This decorator should wrap

/validate/check

Parameters:

• request (Request Object) – The request that is intercepted during the API call
• action (basestring) – An optional Action

Returns:

Always true. Modified the parameter request

privacyidea.api.lib.prepolicy.mock_fail(req, action)

This is a mock function as an example for check_external. This function creates a problem situation and the token/init or token/assign will show this exception accordingly.

privacyidea.api.lib.prepolicy.mock_success(req, action)

This is a mock function as an example for check_external. This function returns success and the API call will go on unmodified.

privacyidea.api.lib.prepolicy.papertoken_count(request=None, action=None)

This is a token specific wrapper for paper token for the endpoint /token/init. According to the policy scope=SCOPE.ENROLL, action=PAPERACTION.PAPER_COUNT it sets the parameter papertoken_count to enroll a paper token with such many OTP values.

Parameters:

• request –
• action –

Returns:

class privacyidea.api.lib.prepolicy.prepolicy(function, request, action=None)

This is the decorator wrapper to call a specific function before an API call. The prepolicy decorator is to be used in the API calls. A prepolicy decorator then will modify the request data or raise an exception

privacyidea.api.lib.prepolicy.realmadmin(request=None, action=None)

This decorator adds the first REALM to the parameters if the administrator, calling this API is a realm admin. This way, if the admin calls e.g. GET /user without realm parameter, he will not see all users, but only users in one of his realms.

TODO: If a realm admin is allowed to see more than one realm,

this is not handled at the moment. We need to change the underlying library functions!

Parameters:

• request – The HTTP reqeust
• action – The action like ACTION.USERLIST

privacyidea.api.lib.prepolicy.required_email(request=None, action=None)

This precondition checks if the “email” parameter matches the regular expression in the policy scope=register, action=requiredemail. See requiredemail.

Check ACTION.REQUIREDEMAIL

This decorator should wrap POST /register

Parameters:

• request – The Request Object
• action – An optional Action

Returns:

Modifies the request parameters or raises an Exception

privacyidea.api.lib.prepolicy.save_client_application_type(request, action)

This decorator is used to write the client IP and the HTTP user agent ( clienttype) to the database.

In fact this is not a policy decorator, as it checks no policy. In fact, we could however one day define this as a policy, too. :param req: :return:

privacyidea.api.lib.prepolicy.set_realm(request=None, action=None)

Pre Policy This pre condition gets the current realm and verifies if the realm should be rewritten due to the policy definition. I takes the realm from the request and - if a policy matches - replaces this realm with the realm defined in the policy

Check ACTION.SETREALM

This decorator should wrap

/validate/check

Parameters:

• request (Request Object) – The request that is intercepted during the API call
• action (basestring) – An optional Action

Returns:

Always true. Modified the parameter request

privacyidea.api.lib.prepolicy.twostep_enrollment_activation(request=None, action=None)

This policy function enables the two-step enrollment process according to the configured policies. It is used to decorate the /token/init endpoint.

If a <type>_2step policy matches, the 2stepinit parameter is handled according to the policy. If no policy matches, the 2stepinit parameter is removed from the request data.

privacyidea.api.lib.prepolicy.twostep_enrollment_parameters(request=None, action=None)

If the 2stepinit parameter is set to true, this policy function reads additional configuration from policies and adds it to request.all_data, that is:

• {type}_2step_serversize is written to 2step_serversize
• {type}_2step_clientsize is written to ``2step_clientsize`
• {type}_2step_difficulty is written to 2step_difficulty

If no policy matches, the value passed by the user is kept.

This policy function is used to decorate the /token/init endpoint.

privacyidea.api.lib.prepolicy.u2ftoken_allowed(request, action)

This is a token specific wrapper for u2f token for the endpoint

/token/init. According to the policy scope=SCOPE.ENROLL, action=U2FACTION.REQ it checks, if the assertion certificate is an allowed U2F token type.

If the token, which is enrolled contains a non allowed attestation certificate, we bail out.

Parameters:

• request –
• action –

Returns:

privacyidea.api.lib.prepolicy.u2ftoken_verify_cert(request, action)

This is a token specific wrapper for u2f token for the endpoint /token/init According to the policy scope=SCOPE.ENROLL, action=U2FACTION.NO_VERIFY_CERT it can add a parameter to the enrollment parameters to not verify the attestation certificate. The default is to verify the cert. :param request: :param action: :return:

Post Policies

These are the policy decorators as POST conditions for the API calls. I.e. these conditions are executed after the wrapped API call. This module uses the policy base functions from privacyidea.lib.policy but also components from flask like g.

Wrapping the functions in a decorator class enables easy modular testing.

The functions of this module are tested in tests/test_api_lib_policy.py

privacyidea.api.lib.postpolicy.add_user_detail_to_response(request, response)

This policy decorated is used in the AUTHZ scope. If the boolean value add_user_in_response is set, the details will contain a dictionary “user” with all user details.

Parameters:

• request –
• response –

Returns:

privacyidea.api.lib.postpolicy.autoassign(request, response)

This decorator decorates the function /validate/check. Depending on ACTION.AUTOASSIGN it checks if the user has no token and if the given OTP-value matches a token in the users realm, that is not yet assigned to any user.

If a token can be found, it assigns the token to the user also taking into account ACTION.MAXTOKENUSER and ACTION.MAXTOKENREALM. :return:

privacyidea.api.lib.postpolicy.check_serial(request, response)

This policy function is to be used in a decorator of an API function. It checks, if the token, that was used in the API call has a serial number that is allowed to be used.

If not, a PolicyException is raised.

Parameters:response (Response object) – The response of the decorated function Returns:A new (maybe modified) response

privacyidea.api.lib.postpolicy.check_tokeninfo(request, response)

This policy function is used as a decorator for the validate API. It checks after a successful authentication if the token has a matching tokeninfo field. If it does not match, authorization is denied. Then a PolicyException is raised.

Parameters:response (Response object) – The response of the decorated function Returns:A new modified response

privacyidea.api.lib.postpolicy.check_tokentype(request, response)

This policy function is to be used in a decorator of an API function. It checks, if the token, that was used in the API call is of a type that is allowed to be used.

If not, a PolicyException is raised.

Parameters:response (Response object) – The response of the decorated function Returns:A new (maybe modified) response

privacyidea.api.lib.postpolicy.construct_radius_response(request, response)

This decorator implements the /validate/radiuscheck endpoint. In case this URL was requested, a successful authentication results in an empty response with a HTTP 204 status code. An unsuccessful authentication results in an empty response with a HTTP 400 status code. :return:

privacyidea.api.lib.postpolicy.get_webui_settings(request, response)

This decorator is used in the /auth API to add configuration information like the logout_time or the policy_template_url to the response. :param request: flask request object :param response: flask response object :return: the response

privacyidea.api.lib.postpolicy.no_detail_on_fail(request, response)

This policy function is used with the AUTHZ scope. If the boolean value no_detail_on_fail is set, the details will be stripped if the authentication request failed.

Parameters:

• request –
• response –

Returns:

privacyidea.api.lib.postpolicy.no_detail_on_success(request, response)

This policy function is used with the AUTHZ scope. If the boolean value no_detail_on_success is set, the details will be stripped if the authentication request was successful.

Parameters:

• request –
• response –

Returns:

privacyidea.api.lib.postpolicy.offline_info(request, response)

This decorator is used with the function /validate/check. It is not triggered by an ordinary policy but by a MachineToken definition. If for the given Client and Token an offline application is defined, the response is enhanced with the offline information - the hashes of the OTP.

class privacyidea.api.lib.postpolicy.postpolicy(function, request=None)

Decorator that allows one to call a specific function after the decorated function. The postpolicy decorator is to be used in the API calls.

class privacyidea.api.lib.postpolicy.postrequest(function, request=None)

Decorator that is supposed to be used with after_request.

privacyidea.api.lib.postpolicy.save_pin_change(request, response, serial=None)

This policy function checks if the next_pin_change date should be stored in the tokeninfo table.

1. Check scope:enrollment and ACTION.CHANGE_PIN_FIRST_USE. This action is used, when the administrator enrolls a token or sets a PIN
2. Check scope:enrollment and ACTION.CHANGE_PIN_EVERY is used, if the user changes the PIN.

This function decorates /token/init and /token/setpin. The parameter “pin” and “otppin” is investigated.

Parameters:

• request –
• action –

Returns:

privacyidea.api.lib.postpolicy.sign_response(request, response)

This decorator is used to sign the response. It adds the nonce from the request, if it exist and adds the nonce and the signature to the response.

Note

This only works for JSON responses. So if we fail to decode the JSON, we just pass on.

The usual way to use it is, to wrap the after_request, so that we can also sign errors.

@postrequest(sign_response, request=request) def after_request(response):

Parameters:

• request – The Request object
• response – The Response object

Policy Decorators

These are the policy decorator functions for internal (lib) policy decorators. policy decorators for the API (pre/post) are defined in api/lib/policy

The functions of this module are tested in tests/test_lib_policy_decorator.py

privacyidea.lib.policydecorators.auth_cache(wrapped_function, user_object, passw, options=None)

Decorate lib.token:check_user_pass. Verify, if the authentication can be found in the auth_cache.

Parameters:

• wrapped_function – usually “check_user_pass”
• user_object – User who tries to authenticate
• passw – The PIN and OTP
• options – Dict containing values for “g” and “clientip”.

Returns:

Tuple of True/False and reply-dictionary

privacyidea.lib.policydecorators.auth_lastauth(wrapped_function, user_or_serial, passw, options=None)

This decorator checks the policy settings of ACTION.LASTAUTH If the last authentication stored in tokeninfo last_auth_success of a token is exceeded, the authentication is denied.

The wrapped function is usually token.check_user_pass, which takes the arguments (user, passw, options={}) OR token.check_serial_pass with the arguments (user, passw, options={})

Parameters:

• wrapped_function – either check_user_pass or check_serial_pass
• user_or_serial – either the User user_or_serial or a serial
• passw –
• options – Dict containing values for “g” and “clientip”

Returns:

Tuple of True/False and reply-dictionary

privacyidea.lib.policydecorators.auth_otppin(wrapped_function, *args, **kwds)

Decorator to decorate the tokenclass.check_pin function. Depending on the ACTION.OTPPIN it * either simply accepts an empty pin * checks the pin against the userstore * or passes the request to the wrapped_function

Parameters:wrapped_function – In this case the wrapped function should be

tokenclass.check_ping :param *args: args[1] is the pin :param **kwds: kwds[“options”] contains the flask g :return: True or False

privacyidea.lib.policydecorators.auth_user_does_not_exist(wrapped_function, user_object, passw, options=None)

This decorator checks, if the user does exist at all. If the user does exist, the wrapped function is called.

The wrapped function is usually token.check_user_pass, which takes the arguments (user, passw, options={})

Parameters:

• wrapped_function –
• user_object –
• passw –
• options – Dict containing values for “g” and “clientip”

Returns:

Tuple of True/False and reply-dictionary

privacyidea.lib.policydecorators.auth_user_has_no_token(wrapped_function, user_object, passw, options=None)

This decorator checks if the user has a token at all. If the user has a token, the wrapped function is called.

The wrapped function is usually token.check_user_pass, which takes the arguments (user, passw, options={})

Parameters:

• wrapped_function –
• user_object –
• passw –
• options – Dict containing values for “g” and “clientip”

Returns:

Tuple of True/False and reply-dictionary

privacyidea.lib.policydecorators.auth_user_passthru(wrapped_function, user_object, passw, options=None)

This decorator checks the policy settings of ACTION.PASSTHRU. If the authentication against the userstore is not successful, the wrapped function is called.

The wrapped function is usually token.check_user_pass, which takes the arguments (user, passw, options={})

Parameters:

• wrapped_function –
• user_object –
• passw –
• options – Dict containing values for “g” and “clientip”

Returns:

Tuple of True/False and reply-dictionary

privacyidea.lib.policydecorators.auth_user_timelimit(wrapped_function, user_object, passw, options=None)

This decorator checks the policy settings of ACTION.AUTHMAXSUCCESS, ACTION.AUTHMAXFAIL If the authentication was successful, it checks, if the number of allowed successful authentications is exceeded (AUTHMAXSUCCESS).

If the AUTHMAXFAIL is exceed it denies even a successful authentication.

The wrapped function is usually token.check_user_pass, which takes the arguments (user, passw, options={})

Parameters:

• wrapped_function –
• user_object –
• passw –
• options – Dict containing values for “g” and “clientip”

Returns:

Tuple of True/False and reply-dictionary

privacyidea.lib.policydecorators.challenge_response_allowed(func)

This decorator is used to wrap tokenclass.is_challenge_request. It checks, if a challenge response authentication is allowed for this token type. To allow this, the policy

scope:authentication, action:challenge_response must be set.

If the tokentype is not allowed for challenge_response, this decorator returns false.

See challenge_response.

Parameters:func – wrapped function

privacyidea.lib.policydecorators.config_lost_token(wrapped_function, *args, **kwds)

Decorator to decorate the lib.token.lost_token function. Depending on ACTION.LOSTTOKENVALID, ACTION.LOSTTOKENPWCONTENTS, ACTION.LOSTTOKENPWLEN it sets the check_otp parameter, to signal how the lostToken should be generated.

Parameters:

• wrapped_function – Usually the function lost_token()
• args – argument “serial” as the old serial number
• kwds – keyword arguments like “validity”, “contents”, “pw_len”

kwds[“options”] contains the flask g

Returns:calls the original function with the modified “validity”,

“contents” and “pw_len” argument

class privacyidea.lib.policydecorators.libpolicy(decorator_function)

This is the decorator wrapper to call a specific function before a library call in contrast to prepolicy and postpolicy, which are to be called in API Calls.

The decorator expects a named parameter “options”. In this options dict it will look for the flask global “g”.

privacyidea.lib.policydecorators.login_mode(wrapped_function, *args, **kwds)

Decorator to decorate the lib.auth.check_webui_user function. Depending on ACTION.LOGINMODE it sets the check_otp parameter, to signal that the authentication should be performed against privacyIDEA.

Parameters:

• wrapped_function – Usually the function check_webui_user
• args – arguments user_obj and password
• kwds – keyword arguments like options and !check_otp!

kwds[“options”] contains the flask g :return: calls the original function with the modified “check_otp” argument

Event Handler

The following event handlers are known to privacyIDEA

Event Handler Base Class

class privacyidea.lib.eventhandler.base.BaseEventHandler

An Eventhandler needs to return a list of actions, which it can handle.

It also returns a list of allowed action and conditions

It returns an identifier, which can be used in the eventhandlig definitions

actions

This method returns a list of available actions, that are provided by this event handler. :return: dictionary of actions.

check_condition(options)

Check if all conditions are met and if the action should be executed. The the conditions are met, we return “True” :return: True

conditions

The UserNotification can filter for conditions like * type of logged in user and * successful or failed value.success

allowed types are str, multi, text, regexp

Returns:dict

description = 'This is the base class of an EventHandler with no functionality'

do(action, options=None)

This method executes the defined action in the given event.

Parameters:

• action –
• options (dict) – Contains the flask parameters g and request and the handler_def configuration

Returns:

events

This method returns a list allowed events, that this event handler can be bound to and which it can handle with the corresponding actions.

An eventhandler may return an asterisk [“*”] indicating, that it can be used in all events. :return: list of events

identifier = 'BaseEventHandler'

User Notification Event Handler

class privacyidea.lib.eventhandler.usernotification.UserNotificationEventHandler

An Eventhandler needs to return a list of actions, which it can handle.

It also returns a list of allowed action and conditions

It returns an identifier, which can be used in the eventhandlig definitions

actions

This method returns a dictionary of allowed actions and possible options in this handler module.

Returns:dict with actions

description = 'This eventhandler notifies the user about actions on his tokens'

do(action, options=None)

This method executes the defined action in the given event.

Parameters:

• action –
• options (dict) – Contains the flask parameters g, request, response and the handler_def configuration

Returns:

identifier = 'UserNotification'

class privacyidea.lib.event.EventConfiguration

This class is supposed to contain the event handling configuration during the Request. It can be read initially (in the init method) an can be accessed later during the request.

events

get_event(eventid)

Return the reduced list with the given eventid. This list should only have one element.

Parameters:eventid (int) – id of the event Returns:list with one element

get_handled_events(eventname)

Return a list of the event handling definitions for the given eventname

Parameters:eventname – Returns:

privacyidea.lib.event.delete_event(event_id)

Delete the event configuration with this given ID. :param event_id: The database ID of the event. :type event_id: int :return:

privacyidea.lib.event.enable_event(event_id, enable=True)

Enable or disable the and event :param event_id: ID of the event :return:

class privacyidea.lib.event.event(eventname, request, g)

This is the event decorator that calls the event handler in the handler module. This event decorator can be used at any API call

privacyidea.lib.event.get_handler_object(handlername)

Return an event handler object based on the Name of the event handler class

Parameters:handlername – The identifier of the Handler Class Returns:

privacyidea.lib.event.set_event(name, event, handlermodule, action, conditions=None, ordering=0, options=None, id=None, active=True)

Set an event handling configuration. This writes an entry to the database eventhandler.

Parameters:

• name – The name of the event definition
• event (basestring) – The name of the event to react on. Can be a single event or a comma separated list.
• handlermodule (basestring) – The identifier of the event handler module. This is an identifier string like “UserNotification”
• action (basestring) – The action to perform. This is an action defined by the handler module
• conditions (dict) – A condition. Only if this condition is met, the action is performed.
• ordering (integer) – An optional ordering of the event definitions.
• options (dict) – Additional options, that are needed as parameters for the action
• id (int) – The DB id of the event. If the id is given, the event is updated. Otherwise a new entry is generated.

Returns:

The id of the event.

SMS Provider

The following SMS providers are know to privacyIDEA

HTTP SMS Provider

class privacyidea.lib.smsprovider.HttpSMSProvider.HttpSMSProvider(db_smsprovider_object=None, smsgateway=None)

classmethod parameters()

Return a dictionary, that describes the parameters and options for the SMS provider. Parameters are required keys to values.

Returns:dict

submit_message(phone, message)

send a message to a phone via an http sms gateway

Parameters:

• phone – the phone number
• message – the message to submit to the phone

Returns:

Sipgate SMS Provider

class privacyidea.lib.smsprovider.SipgateSMSProvider.SipgateSMSProvider(db_smsprovider_object=None, smsgateway=None)

classmethod parameters()

Return a dictionary, that describes the parameters and options for the SMS provider. Parameters are required keys to values.

Returns:dict

submit_message(phone, message)

SMTP SMS Provider

class privacyidea.lib.smsprovider.SmtpSMSProvider.SmtpSMSProvider(db_smsprovider_object=None, smsgateway=None)

classmethod parameters()

Return a dictionary, that describes the parameters and options for the SMS provider. Parameters are required keys to values.

Returns:dict

submit_message(phone, message)

Submits the message for phone to the email gateway.

Returns true in case of success

In case of a failure an exception is raised

SMSProvider is the base class for submitting SMS. It provides 3 different imlementations:

• HTTP: submitting SMS via an HTTP gateway of an SMS provider
• SMTP: submitting SMS via an SMTP gateway of an SMS provider
• Sipgate: submitting SMS via Sipgate service

Base Class

class privacyidea.lib.smsprovider.SMSProvider.ISMSProvider(db_smsprovider_object=None, smsgateway=None)

the SMS Provider Interface - BaseClass

load_config(config_dict)

Load the configuration dictionary

Parameters:config_dict (dict) – The conifugration of the SMS provider Returns:None

classmethod parameters()

Return a dictionary, that describes the parameters and options for the SMS provider. Parameters are required keys to values with defined keys, while options can be any combination.

Each option is the key to another dict, that describes this option, if it is required, a description and which values it can take. The values are optional.

Additional options can not be named in advance. E.g. some provider specific HTTP parameters of HTTP gateways are options. The HTTP parameter for the SMS text could be “text” at one provider and “sms” at another one.

The options can be fixed values or also take the tags {otp}, {user}, {phone}.

Returns:dict

submit_message(phone, message)

Sends the SMS. It should return a bool indicating if the SMS was sent successfully.

In case of SMS send fail, an Exception should be raised. :return: Success :rtype: bool

UserIdResolvers

The useridresolver is responsible for getting userids for loginnames and vice versa.

This base module contains the base class UserIdResolver.UserIdResolver and also the community class PasswdIdResolver.IdResolver, that is inherited from the base class.

Base class

class privacyidea.lib.resolvers.UserIdResolver.UserIdResolver

add_user(attributes=None)

Add a new user in the useridresolver. This is only possible, if the UserIdResolver supports this and if we have write access to the user store.

Parameters:

• username (basestring) – The login name of the user
• attributes – Attributes according to the attribute mapping

Returns:

The new UID of the user. The UserIdResolver needs to

determine the way how to create the UID.

checkPass(uid, password)

This function checks the password for a given uid. returns true in case of success false if password does not match

Parameters:

• uid (string or int) – The uid in the resolver
• password (string) – the password to check. Usually in cleartext

Returns:

True or False

Return type:

bool

close()

Hook to close down the resolver after one request

delete_user(uid)

Delete a user from the useridresolver. The user is referenced by the user id. :param uid: The uid of the user object, that should be deleted. :type uid: basestring :return: Returns True in case of success :rtype: bool

editable

Return true, if the Instance! of this resolver is configured editable. :return:

classmethod getResolverClassDescriptor()

return the descriptor of the resolver, which is - the class name and - the config description

Returns:resolver description dict Return type:dict

static getResolverClassType()

provide the resolver type for registration

static getResolverDescriptor()

return the descriptor of the resolver, which is - the class name and - the config description

Returns:resolver description dict Return type:dict

getResolverId()

get resolver specific information :return: the resolver identifier string - empty string if not exist

static getResolverType()

getResolverType - return the type of the resolver

Returns:returns the string ‘ldapresolver’ Return type:string

getUserId(loginName)

The loginname is resolved to a user_id. Depending on the resolver type the user_id can be an ID (like in /etc/passwd) or a string (like the DN in LDAP)

It needs to return an emptry string, if the user does not exist.

Parameters:loginName (sting) – The login name of the user Returns:The ID of the user Return type:string or int

getUserInfo(userid)

This function returns all user information for a given user object identified by UserID. :param userid: ID of the user in the resolver :type userid: int or string :return: dictionary, if no object is found, the dictionary is empty :rtype: dict

getUserList(searchDict=None)

This function finds the user objects, that have the term ‘value’ in the user object field ‘key’

Parameters:searchDict (dict) – dict with key values of user attributes - the key may be something like ‘loginname’ or ‘email’ the value is a regular expression. Returns:list of dictionaries (each dictionary contains a user object) or an empty string if no object is found. Return type:list of dicts

getUsername(userid)

Returns the username/loginname for a given userid :param userid: The userid in this resolver :type userid: string :return: username :rtype: string

loadConfig(config)

Load the configuration from the dict into the Resolver object. If attributes are missing, need to set default values. If required attributes are missing, this should raise an Exception.

Parameters:config (dict) – The configuration values of the resolver

static testconnection(param)

This function lets you test if the parameters can be used to create a working resolver. The implementation should try to connect to the user store and verify if users can be retrieved. In case of success it should return a text like “Resolver config seems OK. 123 Users found.”

param param: The parameters that should be saved as the resolver type param: dict return: returns True in case of success and a descriptive text rtype: tuple

update_user(uid, attributes=None)

Update an existing user. This function is also used to update the password. Since the attribute mapping know, which field contains the password, this function can also take care for password changing.

Attributes that are not contained in the dict attributes are not modified.

Parameters:

• uid (basestring) – The uid of the user object in the resolver.
• attributes (dict) – Attributes to be updated.

Returns:

True in case of success

PasswdResolver

class privacyidea.lib.resolvers.PasswdIdResolver.IdResolver

checkPass(uid, password)

This function checks the password for a given uid. returns true in case of success false if password does not match

We do not support shadow passwords. so the seconds column of the passwd file needs to contain the crypted password

If the password is a unicode object, it is encoded according to ENCODING first.

Parameters:

• uid (int) – The uid of the user
• password (sting) – The password in cleartext

Returns:

True or False

Return type:

bool

checkUserId(line, pattern)

Check if a userid matches a pattern. A pattern can be “=1000”, “>=1000”, “<2000” or “between 1000,2000”.

Parameters:

• line (dict) – the dictionary of a user
• pattern (string) – match pattern with <, <=...

Returns:

True or False

Return type:

bool

checkUserName(line, pattern)

check for user name

classmethod getResolverClassDescriptor()

return the descriptor of the resolver, which is - the class name and - the config description

Returns:resolver description dict Return type:dict

getResolverId()

return the resolver identifier string, which in fact is filename, where it points to.

getSearchFields(searchDict=None)

show, which search fields this userIdResolver supports

TODO: implementation is not completed

Parameters:searchDict (dict) – fields, which can be queried Returns:dict of all searchFields Return type:dict

getUserId(LoginName)

search the user id from the login name

Parameters:LoginName – the login of the user (as unicode) Returns:the userId

getUserInfo(userId, no_passwd=False)

get some info about the user as we only have the loginId, we have to traverse the dict for the value

Parameters:

• userId – the to be searched user
• no_passwd – retrun no password

Returns:

dict of user info

getUserList(searchDict)

get a list of all users matching the search criteria of the searchdict

Parameters:searchDict – dict of search expressions

getUsername(userId)

Returns the username/loginname for a given userid :param userid: The userid in this resolver :type userid: string :return: username :rtype: string

loadConfig(configDict)

The UserIdResolver could be configured from the pylons app config - here this could be the passwd file , whether it is /etc/passwd or /etc/shadow

loadFile()

Loads the data of the file initially. if the self.fileName is empty, it loads /etc/passwd. Empty lines are ignored.

static setup(config=None, cache_dir=None)

this setup hook is triggered, when the server starts to serve the first request

Parameters:config (the privacyidea config dict) – the privacyidea config

LDAPResolver

class privacyidea.lib.resolvers.LDAPIdResolver.IdResolver

add_user(attributes=None)

Add a new user to the LDAP directory. The user can only be created in the LDAP using a DN. So we have to construct the DN out of the given attributes.

attributes are these “username”, “surname”, “givenname”, “email”, “mobile”, “phone”, “password”

Parameters:attributes (dict) – Attributes according to the attribute mapping Returns:The new UID of the user. The UserIdResolver needs to

determine the way how to create the UID.

checkPass(uid, password)

This function checks the password for a given uid. - returns true in case of success - false if password does not match

static create_connection(authtype=None, server=None, user=None, password=None, auto_bind=False, client_strategy='SYNC', check_names=True, auto_referrals=False, receive_timeout=5, start_tls=False)

Create a connection to the LDAP server.

Parameters:

• authtype –
• server –
• user –
• password –
• auto_bind –
• client_strategy –
• check_names –
• auto_referrals –
• receive_timeout – At the moment we do not use this, since receive_timeout is not supported by ldap3 < 2.

Returns:

delete_user(uid)

Delete a user from the LDAP Directory.

The user is referenced by the user id. :param uid: The uid of the user object, that should be deleted. :type uid: basestring :return: Returns True in case of success :rtype: bool

editable

Return true, if the instance of the resolver is configured editable :return:

classmethod getResolverClassDescriptor()

return the descriptor of the resolver, which is - the class name and - the config description

Returns:resolver description dict Return type:dict

getResolverId()

Returns the resolver Id This should be an Identifier of the resolver, preferable the type and the name of the resolver.

getUserId(LoginName)

resolve the loginname to the userid.

Parameters:LoginName (string) – The login name from the credentials Returns:UserId as found for the LoginName

getUserInfo(userId)

This function returns all user info for a given userid/object.

Parameters:userId (string) – The userid of the object Returns:A dictionary with the keys defined in self.userinfo Return type:dict

getUserList(searchDict)

Parameters:searchDict (dict) – A dictionary with search parameters Returns:list of users, where each user is a dictionary

getUsername(user_id)

Returns the username/loginname for a given user_id :param user_id: The user_id in this resolver :type user_id: string :return: username :rtype: string

classmethod get_serverpool(urilist, timeout, get_info=None, tls_context=None, rounds=2, exhaust=30)

This create the serverpool for the ldap3 connection. The URI from the LDAP resolver can contain a comma separated list of LDAP servers. These are split and then added to the pool.

See https://github.com/cannatag/ldap3/blob/master/docs/manual/source/servers.rst#server-pool

Parameters:

• urilist (basestring) – The list of LDAP URIs, comma separated
• timeout (float) – The connection timeout
• get_info – The get_info type passed to the ldap3.Sever constructor. default: ldap3.SCHEMA, should be ldap3.NONE in case of a bind.
• tls_context – A ldap3.tls object, which defines if certificate verification should be performed
• rounds – The number of rounds we should cycle through the server pool before giving up
• exhaust – The seconds, for how long a non-reachable server should be removed from the serverpool

Returns:

Server Pool

Return type:

LDAP3 Server Pool Instance

loadConfig(config)

Load the config from conf.

Parameters:config (dict) – The configuration from the Config Table

‘#ldap_uri’: ‘LDAPURI’, ‘#ldap_basedn’: ‘LDAPBASE’, ‘#ldap_binddn’: ‘BINDDN’, ‘#ldap_password’: ‘BINDPW’, ‘#ldap_timeout’: ‘TIMEOUT’, ‘#ldap_sizelimit’: ‘SIZELIMIT’, ‘#ldap_loginattr’: ‘LOGINNAMEATTRIBUTE’, ‘#ldap_searchfilter’: ‘LDAPSEARCHFILTER’, ‘#ldap_mapping’: ‘USERINFO’, ‘#ldap_uidtype’: ‘UIDTYPE’, ‘#ldap_noreferrals’ : ‘NOREFERRALS’, ‘#ldap_editable’ : ‘EDITABLE’, ‘#ldap_certificate’: ‘CACERTIFICATE’,

static split_uri(uri)

Splits LDAP URIs like: * ldap://server * ldaps://server * ldap[s]://server:1234 * server :param uri: The LDAP URI :return: Returns a tuple of Servername, Port and SSL(bool)

classmethod testconnection(param)

This function lets you test the to be saved LDAP connection.

Parameters:param (dict) – A dictionary with all necessary parameter to test the connection. Returns:Tuple of success and a description Return type:(bool, string)

Parameters are:

BINDDN, BINDPW, LDAPURI, TIMEOUT, LDAPBASE, LOGINNAMEATTRIBUTE, LDAPSEARCHFILTER, USERINFO, SIZELIMIT, NOREFERRALS, CACERTIFICATE, AUTHTYPE, TLS_VERIFY, TLS_VERSION, TLS_CA_FILE, SERVERPOOL_ROUNDS, SERVERPOOL_SKIP

update_user(uid, attributes=None)

Update an existing user. This function is also used to update the password. Since the attribute mapping know, which field contains the password, this function can also take care for password changing.

Attributes that are not contained in the dict attributes are not modified.

Parameters:

• uid (basestring) – The uid of the user object in the resolver.
• attributes (dict) – Attributes to be updated.

Returns:

True in case of success

Audit log

Base class

class privacyidea.lib.auditmodules.base.Audit(config=None)

add_to_log(param)

Add to existing log entry :param param: :return:

audit_entry_to_dict(audit_entry)

If the search_query returns an iterator with elements that are not a dictionary, the audit module needs to provide this function, to convert the audit entry to a dictionary.

csv_generator(param=None, user=None, timelimit=None)

A generator that can be used to stream the audit log

Parameters:param – Returns:

finalize_log()

This method is called to finalize the audit_data. I.e. sign the data and write it to the database. It should hash the data and do a hash chain and sign the data

get_audit_id()

get_count(search_dict, timedelta=None, success=None)

Returns the number of found log entries. E.g. used for checking the timelimit.

Parameters:param – List of filter parameters Returns:number of found entries

get_dataframe(start_time=datetime.datetime(2018, 8, 2, 9, 30, 34, 818959), end_time=datetime.datetime(2018, 8, 9, 9, 30, 34, 819189))

The Audit module can handle its data the best. This function is used to return a pandas.dataframe with all audit data in the given time frame.

This dataframe then can be used for extracting statistics.

Parameters:

• start_time (datetime) – The start time of the data
• end_time (datetime) – The end time of the data

Returns:

Audit data

Return type:

dataframe

get_total(param, AND=True, display_error=True)

This method returns the total number of audit entries in the audit store

initialize()

initialize_log(param)

This method initialized the log state. The fact, that the log state was initialized, also needs to be logged. Therefor the same params are passed as i the log method.

log(param)

This method is used to log the data. During a request this method can be called several times to fill the internal audit_data dictionary.

log_token_num(count)

Log the number of the tokens. Can be passed like log_token_num(get_tokens(count=True))

Parameters:count (int) – Number of tokens Returns:

read_keys(pub, priv)

Set the private and public key for the audit class. This is achieved by passing the entries.

#priv = config.get(“privacyideaAudit.key.private”) #pub = config.get(“privacyideaAudit.key.public”)

Parameters:

• pub (string with filename) – Public key, used for verifying the signature
• priv (string with filename) – Private key, used to sign the audit entry

Returns:

None

search(param, display_error=True, rp_dict=None, timelimit=None)

This function is used to search audit events.

param: Search parameters can be passed.

return: A pagination object

This function is deprecated.

search_query(search_dict, rp_dict)

This function returns the audit log as an iterator on the result

SQL Audit module

class privacyidea.lib.auditmodules.sqlaudit.Audit(config=None)

This is the SQLAudit module, which writes the audit entries to an SQL database table. It requires the configuration parameters. PI_AUDIT_SQL_URI

add_to_log(param)

Add new text to an existing log entry :param param: :return:

clear()

Deletes all entries in the database table. This is only used for test cases! :return:

csv_generator(param=None, user=None, timelimit=None)

Returns the audit log as csv file. :param config: The current flask app configuration :type config: dict :param param: The request parameters :type param: dict :param user: The user, who issued the request :return: None. It yields results as a generator

finalize_log()

This method is used to log the data. It should hash the data and do a hash chain and sign the data

get_dataframe(start_time=datetime.datetime(2018, 8, 2, 9, 30, 35, 238128), end_time=datetime.datetime(2018, 8, 9, 9, 30, 35, 238236))

The Audit module can handle its data the best. This function is used to return a pandas.dataframe with all audit data in the given time frame.

This dataframe then can be used for extracting statistics.

Parameters:

• start_time (datetime) – The start time of the data
• end_time (datetime) – The end time of the data

Returns:

Audit data

Return type:

dataframe

get_total(param, AND=True, display_error=True, timelimit=None)

This method returns the total number of audit entries in the audit store

log(param)

Add new log details in param to the internal log data self.audit_data.

Parameters:param (dict) – Log data that is to be added Returns:None

read_keys(pub, priv)

Set the private and public key for the audit class. This is achieved by passing the entries.

#priv = config.get(“privacyideaAudit.key.private”) #pub = config.get(“privacyideaAudit.key.public”)

Parameters:

• pub (string with filename) – Public key, used for verifying the signature
• priv (string with filename) – Private key, used to sign the audit entry

Returns:

None

search(search_dict, page_size=15, page=1, sortorder='asc', timelimit=None)

This function returns the audit log as a Pagination object.

Parameters:timelimit (timedelta) – Only audit entries newer than this timedelta will be searched

search_query(search_dict, page_size=15, page=1, sortorder='asc', sortname='number', timelimit=None)

This function returns the audit log as an iterator on the result

Parameters:timelimit (timedelta) – Only audit entries newer than this timedelta will be searched

Machine Resolvers

Machine Resolvers are used to find machines in directories like LDAP, Active Directory, puppet, salt, or the /etc/hosts file.

Machines can then be used to assign applications and tokens to those machines.

Base class

class privacyidea.lib.machines.base.BaseMachineResolver(name, config=None)

static get_config_description()

Returns a description what config values are expected and allowed.

Returns:dict

get_machine_id(hostname=None, ip=None)

Returns the machine id for a given hostname or IP address.

If hostname and ip is given, the resolver should also check that the hostname matches the IP. If it can check this and hostname and IP do not match, then an Exception must be raised.

Parameters:

• hostname (basestring) – The hostname of the machine
• ip (netaddr) – IP address of the machine

Returns:

The machine ID, which depends on the resolver

Return type:

basestring

get_machines(machine_id=None, hostname=None, ip=None, any=None, substring=False)

Return a list of all machine objects in this resolver

Parameters:substring – If set to true, it will also match search_hostnames,

that only are a subnet of the machines hostname. :type substring: bool :param any: a substring that matches EITHER hostname, machineid or ip :type any: basestring :return: list of machine objects

load_config(config)

This loads the configuration dictionary, which contains the necessary information for the machine resolver to find and connect to the machine store.

Parameters:config (dict) – The configuration dictionary to run the machine resolver Returns:None

static testconnection(params)

This method can test if the passed parameters would create a working machine resolver.

Parameters:params – Returns:tupple of success and description Return type:(bool, string)

Hosts Machine Resolver

class privacyidea.lib.machines.hosts.HostsMachineResolver(name, config=None)

get_machine_id(hostname=None, ip=None)

Returns the machine id for a given hostname or IP address.

If hostname and ip is given, the resolver should also check that the hostname matches the IP. If it can check this and hostname and IP do not match, then an Exception must be raised.

Parameters:

• hostname (basestring) – The hostname of the machine
• ip (netaddr) – IP address of the machine

Returns:

The machine ID, which depends on the resolver

Return type:

basestring

get_machines(machine_id=None, hostname=None, ip=None, any=None, substring=False)

Return matching machines.

Parameters:

• machine_id – can be matched as substring
• hostname – can be matched as substring
• ip – can not be matched as substring
• substring (bool) – Whether the filtering should be a substring matching
• any (basestring) – a substring that matches EITHER hostname, machineid or ip

Returns:

list of Machine Objects

load_config(config)

This loads the configuration dictionary, which contains the necessary information for the machine resolver to find and connect to the machine store.

Parameters:config (dict) – The configuration dictionary to run the machine resolver Returns:None

static testconnection(params)

Test if the given filename exists.

Parameters:params – Returns:

PinHandler

This module provides the PIN Handling base class. In case of enrolling a token, a PIN Handling class can be used to send the PIN via Email, call an external program or print a letter.

This module is not tested explicitly. It is tested in conjunction with the policy decorator init_random_pin in tests/test_api_lib_policy.py

Base class

class privacyidea.lib.pinhandling.base.PinHandler(options=None)

A PinHandler Class is responsible for handling the OTP PIN during enrollment.

It receives the necessary data like

• the PIN
• the serial number of the token
• the username
• all other user data:
  • given name, surname
  • email address
  • telephone
  • mobile (if the module would deliver via SMS)
• the administrator name (who enrolled the token)

send(pin, serial, user, tokentype=None, logged_in_user=None, userdata=None, options=None)

Parameters:

• pin – The PIN in cleartext
• user (user object) – the owner of the token
• tokentype (basestring) – the type of the token
• logged_in_user (dict) – The logged in user, who enrolled the token
• userdata (dict) – Handler-specific user data like email, mobile...
• options (dict) – Handler-specific additional options

Returns:

True in case of success

Return type:

bool

DB level

On the DB level you can simply modify all objects.

The database model

class privacyidea.models.Admin(**kwargs)

The administrators for managing the system. To manage the administrators use the command pi-manage.

In addition certain realms can be defined to be administrative realms.

Parameters:

• username (basestring) – The username of the admin
• password (basestring) – The password of the admin (stored using PBKDF2, salt and pepper)
• email (basestring) – The email address of the admin (not used at the moment)

class privacyidea.models.Audit(action='', success=0, serial='', token_type='', user='', realm='', resolver='', administrator='', action_detail='', info='', privacyidea_server='', client='', loglevel='default', clearance_level='default')

This class stores the Audit entries

class privacyidea.models.CAConnector(name, catype)

The table “caconnector” contains the names and types of the defined CA connectors. Each connector has a different configuration, that is stored in the table “caconnectorconfig”.

class privacyidea.models.CAConnectorConfig(caconnector_id=None, Key=None, Value=None, caconnector=None, Type='', Description='')

Each CAConnector can have multiple configuration entries. Each CA Connector type can have different required config values. Therefor the configuration is stored in simple key/value pairs. If the type of a config entry is set to “password” the value of this config entry is stored encrypted.

The config entries are referenced by the id of the resolver.

class privacyidea.models.Challenge(serial, transaction_id=None, challenge=u'', data=u'', session=u'', validitytime=120)

Table for handling of the generic challenges.

get(timestamp=False)

return a dictionary of all vars in the challenge class

Parameters:timestamp (bool) – if true, the timestamp will given in a readable format 2014-11-29 21:56:43.057293 Returns:dict of vars

get_otp_status()

This returns how many OTPs were already received for this challenge. and if a valid OTP was received.

Returns:tuple of count and True/False Return type:tuple

is_valid()

Returns true, if the expiration time has not passed, yet. :return: True if valid :rtype: bool

set_data(data)

set the internal data of the challenge :param data: unicode data :type data: string, length 512

class privacyidea.models.ClientApplication(**kwargs)

This table stores the clients, which sent an authentication request to privacyIDEA. This table is filled automatically by authentication requests.

class privacyidea.models.Config(Key, Value, Type=u'', Description=u'')

The config table holds all the system configuration in key value pairs.

Additional configuration for realms, resolvers and machine resolvers is stored in specific tables.

class privacyidea.models.EventCounter(name, value=0)

This table stores counters of the event hanlder “Counter”.

increase()

Increase the value of a counter :return:

class privacyidea.models.EventHandler(name, event, handlermodule, action, condition='', ordering=0, options=None, id=None, conditions=None, active=True)

This model holds the list of defined events and actions to this events. A handler module can be bound to an event with the corresponding condition and action.

get()

Return the serialized policy object including the options

Returns:complete dict Rytpe:dict

class privacyidea.models.EventHandlerCondition(eventhandler_id, Key, Value, comparator='equal')

Each EventHandler entry can have additional conditions according to the handler module

class privacyidea.models.EventHandlerOption(eventhandler_id, Key, Value, Type='', Description='')

Each EventHandler entry can have additional options according to the handler module.

class privacyidea.models.MachineResolver(name, rtype)

This model holds the definition to the machinestore. Machines could be located in flat files, LDAP directory or in puppet services or other...

The usual MachineResolver just holds a name and a type and a reference to its config

class privacyidea.models.MachineResolverConfig(resolver_id=None, Key=None, Value=None, resolver=None, Type='', Description='')

Each Machine Resolver can have multiple configuration entries. The config entries are referenced by the id of the machine resolver

class privacyidea.models.MachineToken(machineresolver_id=None, machineresolver=None, machine_id=None, token_id=None, serial=None, application=None)

The MachineToken assigns a Token and an application type to a machine. The Machine is represented as the tuple of machineresolver.id and the machine_id. The machine_id is defined by the machineresolver.

This can be an n:m mapping.

class privacyidea.models.MachineTokenOptions(machinetoken_id, key, value)

This class holds an Option for the token assigned to a certain client machine. Each Token-Clientmachine-Combination can have several options.

class privacyidea.models.MethodsMixin

This class mixes in some common Class table functions like delete and save

class privacyidea.models.PasswordReset(recoverycode, username, realm, resolver='', email=None, timestamp=None, expiration=None, expiration_seconds=3600)

Table for handling password resets. This table stores the recoverycodes sent to a given user

The application should save the HASH of the recovery code. Just like the password for the Admins the appliaction shall salt and pepper the hash of the recoverycode. A database admin will not be able to inject a rogue recovery code.

A user can get several recoverycodes. A recovery code has a validity period

Optional: The email to which the recoverycode was sent, can be stored.

class privacyidea.models.Policy(name, active=True, scope='', action='', realm='', adminrealm='', resolver='', user='', client='', time='', condition=0, check_all_resolvers=False)

The policy table contains policy definitions which control the behaviour during

• enrollment
• authentication
• authorization
• administration
• user actions

get(key=None)

Either returns the complete policy entry or a single value :param key: return the value for this key :type key: string :return: complete dict or single value :rytpe: dict or value

class privacyidea.models.PrivacyIDEAServer(**kwargs)

This table can store remote privacyIDEA server definitions

class privacyidea.models.RADIUSServer(**kwargs)

This table can store configurations of RADIUS servers. https://github.com/privacyidea/privacyidea/issues/321

It saves * a unique name * a description * an IP address a * a Port * a secret * timeout in seconds (default 5) * retries (default 3)

These RADIUS server definition can be used in RADIUS tokens or in a radius passthru policy.

save()

If a RADIUS server with a given name is save, then the existing RADIUS server is updated.

class privacyidea.models.Realm(realm)

The realm table contains the defined realms. User Resolvers can be grouped to realms. This very table contains just contains the names of the realms. The linking to resolvers is stored in the table “resolverrealm”.

class privacyidea.models.Resolver(name, rtype)

The table “resolver” contains the names and types of the defined User Resolvers. As each Resolver can have different required config values the configuration of the resolvers is stored in the table “resolverconfig”.

class privacyidea.models.ResolverConfig(resolver_id=None, Key=None, Value=None, resolver=None, Type='', Description='')

Each Resolver can have multiple configuration entries. Each Resolver type can have different required config values. Therefor the configuration is stored in simple key/value pairs. If the type of a config entry is set to “password” the value of this config entry is stored encrypted.

The config entries are referenced by the id of the resolver.

class privacyidea.models.ResolverRealm(resolver_id=None, realm_id=None, resolver_name=None, realm_name=None, priority=None)

This table stores which Resolver is located in which realm This is a N:M relation

class privacyidea.models.SMSGateway(identifier, providermodule, description=None, options=None)

This table stores the SMS Gateway definitions. See https://github.com/privacyidea/privacyidea/wiki/concept:-Delivery-Gateway

It saves the * unique name * a description * the SMS provider module

All options and parameters are saved in other tables.

as_dict()

Return the object as a dictionary

Returns:complete dict Rytpe:dict

delete()

When deleting an SMS Gateway we also delete all the options. :return:

option_dict

Return all connected options as a dictionary

Returns:dict

class privacyidea.models.SMSGatewayOption(gateway_id, Key, Value, Type=None)

This table stores the options and parameters for an SMS Gateway definition.

class privacyidea.models.SMTPServer(**kwargs)

This table can store configurations for SMTP servers. Each entry represents an SMTP server. EMail Token, SMS SMTP Gateways or Notifications like PIN handlers are supposed to use a reference to to a server definition. Each Machine Resolver can have multiple configuration entries. The config entries are referenced by the id of the machine resolver

class privacyidea.models.Subscription(**kwargs)

This table stores the imported subscription files.

get()

Return the database object as dict :return:

class privacyidea.models.TimestampMethodsMixin

This class mixes in the table functions including update of the timestamp

class privacyidea.models.Token(serial, tokentype=u'', isactive=True, otplen=6, otpkey=u'', userid=None, resolver=None, realm=None, **kwargs)

The table “token” contains the basic token data like

• serial number
• assigned user
• secret key...

while the table “tokeninfo” contains additional information that is specific to the tokentype.

del_info(key=None)

Deletes tokeninfo for a given token. If the key is omitted, all Tokeninfo is deleted.

Parameters:key – searches for the given key to delete the entry Returns:

get(key=None, fallback=None, save=False)

simulate the dict behaviour to make challenge processing easier, as this will have to deal as well with ‘dict only challenges’

Parameters:

• key – the attribute name - in case of key is not provided, a dict of all class attributes are returned
• fallback – if the attribute is not found, the fallback is returned
• save – in case of all attributes and save==True, the timestamp is converted to a string representation

get_hashed_pin(pin)

calculate a hash from a pin Fix for working with MS SQL servers MS SQL servers sometimes return a ‘<space>’ when the column is empty: ‘’

get_info()

Returns:The token info as dictionary

get_realms()

return a list of the assigned realms :return: realms :rtype: list

get_user_pin()

return the userPin :rtype : the PIN as a secretObject

set_info(info)

Set the additional token info for this token

Entries that end with ”.type” are used as type for the keys. I.e. two entries sshkey=”XYZ” and sshkey.type=”password” will store the key sshkey as type “password”.

Parameters:info (dict) – The key-values to set for this token

set_pin(pin, hashed=True)

set the OTP pin in a hashed way

set_realms(realms, add=False)

Set the list of the realms. This is done by filling the tokenrealm table. :param realms: realms :type realms: list :param add: If set, the realms are added. I.e. old realms are not

deleted

set_so_pin(soPin)

For smartcards this sets the security officer pin of the token

:rtype : None

split_pin_pass(passwd, prepend=True)

The password is split into the PIN and the OTP component. THe token knows its length, so it can split accordingly.

Parameters:

• passwd – The password that is to be split
• prepend – The PIN is put in front of the OTP value

Returns:

tuple of (res, pin, otpval)

update_otpkey(otpkey)

in case of a new hOtpKey we have to do some more things

update_type(typ)

in case the previous has been different type we must reset the counters But be aware, ray, this could also be upper and lower case mixing...

class privacyidea.models.TokenInfo(token_id, Key, Value, Type=None, Description=None)

The table “tokeninfo” is used to store additional, long information that is specific to the tokentype. E.g. the tokentype “TOTP” has additional entries in the tokeninfo table for “timeStep” and “timeWindow”, which are stored in the column “Key” and “Value”.

The tokeninfo is reference by the foreign key to the “token” table.

class privacyidea.models.TokenRealm(realm_id=0, token_id=0, realmname=None)

This table stored to wich realms a token is assigned. A token is in the realm of the user it is assigned to. But a token can also be put into many additional realms.

save()

We only save this, if it does not exist, yet.

privacyidea.models.cleanup_challenges()

Delete all challenges, that have expired.

Returns:None

privacyidea.models.get_machineresolver_id(resolvername)

Return the database ID of the machine resolver :param resolvername: :return:

privacyidea.models.get_machinetoken_id(machine_id, resolver_name, serial, application)

Returns the ID in the machinetoken table

Parameters:

• machine_id (basestring) – The resolverdependent machine_id
• resolver_name (basestring) – The name of the resolver
• serial (basestring) – the serial number of the token
• application (basestring) – The application type

Returns:

The ID of the machinetoken entry

Return type:

int

privacyidea.models.get_token_id(serial)

Return the database token ID for a given serial number :param serial: :return: token ID :rtpye: int

Frequently Asked Questions

Customization

Templates

You can change the HTML views of the web UI by standard means of the Apache webserver.

All html views are contained in:

static/components/<component>/views/<view>.html

If you want to change the look and feel of the UI, we recommend to define rewrite rules in the webserver. You should create a directory like /etc/privacyidea/customization/ and put your modified views in there. This way you can avoid that your changes get overwritten by a system update.

In the Apache configuration you can add entries like:

RewriteEngine On
RewriteRule "/static/components/login/views/login.html"  \
     "/etc/privacyidea/customization/mylogin.html"

and apply all required changes to the file mylogin.html.

Note

Of course - if there are functional enhancements or bug fixes in the original templates - your template will also not be affected by these.

Themes

You can create your own CSS file to adapt the look and feel of the Web UI. The default CSS is the bootstrap CSS theme. Using PI_CSS you can specify the URL of your own CSS file. The default CSS file url is /static/contrib/css/bootstrap-theme.css. The file in the file system is located at privacyidea/static/contrib/css. You might add a directory privacyidea/static/custom/css/ and add your CSS file there.

A good stating point might be the themes at http://bootswatch.com.

Note

If you add your own CSS file, the file bootstrap-theme.css will not be loaded anymore. So you might start with a copy of the original file.

How can I create users in the privacyIDEA Web UI?

So you installed privacyIDEA and want to enroll tokens to the users and are wondering how to create users.

privacyIDEA can read users from different existing sources like LDAP, SQL, flat files and SCIM.

You very much likely already have an application (like your VPN or a Web Application...) for which you want to increase the logon security. Then this application already knows users. Either in an LDAP or in an SQL database. Most web applications keep their users in a (My)SQL database. And you also need to create users in this very user database for the user to be able to use this application.

Please read the sections UserIdResolvers and Userview for more details.

But you also can define and editable SQL resolver. I.e. you can edit and create new users in an SQL user store.

If you do not have an existing SQL database with users, you can simple create a new database with one table for the users and according rows.

So what’s the thing with all the admins?

privacyIDEA comes with its own admins, who are stored in a database table Admin in its own database (The database model). You can use the tool pi-manage to manage those admins from the command line as the system’s root user. (see Installation)

These admin users can logon to the WebUI using the admin’s user name and the specified password. These admins are used to get a simple quick start.

Then you can define realms (see Realms), that should be administrative realms. I.e. each user in this realm will have administrative rights in the WebUI.

Note

You need to configure these realms within privacyIDEA. Only after these realms exist, you can raise their rights to an administrative role.

Note

Use this carefully. Imagine you defined a resolver to a specific group in your Active Directory to be the pricacyIDEA admins. Then the Active Directory domain admins can simply add users to be administrator in privacyIDEA.

You define the administrative realms in the config file pi.cfg, which is usually located at /etc/privacyidea/pi.cfg:

SUPERUSER_REALM = ["adminrealm1", "super", "boss"]

In this case all the users in the realms “adminrealm1”, “super” and “boss” will have administrative rights in the WebUI, when they login with this realm.

As for all other users, you can use the login_mode to define, if these administrators should login to the WebUI with their userstore password or with an OTP token.

What are possible rollout strategies?

There are different ways to enroll tokens to a big number of users. Here are some selected high level ideas, you can do with privacyIDEA.

Autoenrollment

Using the autoassignment policy you can distribute physical tokens to the users. The users just start using the tokens.

Registration Code

If your users are physically not available and spread around the world, you can send a registration code to the users by postal mail. The registration code is a special token type which can be used by the user to authenticate with 2FA. If used once, the registration token get deleted and can not be used anymore. While logged in, the user can enroll a token on his own.

Automatic initial synchronization

Hardware TOTP tokens may get out of sync due to clock shift. HOTP tokens may get out of sync due to unused keypresses. To cope with this you can activate autosync.

But if you are importing hardware tokens, the clock in the TOTP token may already be out of sync and you do not want the user to authenticate twice, where the first authentication fails.

In this case you can use the following workflow.

In the TOTP token settings you can set the timeWindow to a very high value. Note that this timeWindow are the seconds that privacyIDEA will search for the valid OTP value before and after the current time. E.g. you can set this to 86400. This way you allow the clock in the TOTP token to have drifted for a maximum of one day.

As you do not want such a big window for all authentications, you can automatically reset the timeWindow. You can achieve this by creating an event definition:

• event: validate_check
• handler: token
• condition: * tokentype=TOTP * count_auth_success=1
• action=set tokeninfo * key=*timeWindow* * value=*180*

This way with the first successful authentication of a TOTP token the timeWindow of the TOTP token is set to 180 seconds.

How can I translate to my language?

The web UI can be translated into different languages. The system determines the preferred language of you browser and displays the web UI accordingly.

At the moment “en” and “de” are available.

What are possible migration strategies?

You are already running an OTP system like RSA SecurID, SafeNet or Vasco (alphabetical order) but you would like to migrate to privacyIDEA.

There are different migration strategies using the RADIUS token or the RADIUS passthru policy.

RADIUS token migration strategy

Configure your application like your VPN to authenticate against the privacyIDEA RADIUS server and not against the old deprecated RADIUS server.

Now, you can enroll a RADIUS token for each user, who is supposed to login to this application. Configure the RADIUS token for each user so that the RADIUS request is forwarded to the old RADIUS server.

Now you can start to enroll tokens for the users within privacyIDEA. After enrolling a new token in privacyIDEA you can delete the RADIUS token for this user.

When all RADIUS tokens are deleted, you can switch off the old RADIUS server.

For strategies on enrolling token see What are possible rollout strategies?.

RADIUS PASSTHRU policy migration strategy

Configure your application like your VPN to authenticate against the privacyIDEA RADIUS server and not against the old deprecated RADIUS server.

Starting with privacyIDEA 2.11 the passthru policy was enhanced. You can define a system wide RADIUS server. Then you can create a authentication policy with the passthru action pointing to this RADIUS server. This means that - as long as a user trying to authenticate - has not token assigned, all authentication request with this very username and the password are forwarded to this RADIUS server.

As soon as you enroll a new token for this user in privacyIDEA the user will authenticate with this very token within privacyIDEA an his authentication request will not be forwarded anymore.

As soon as all users have a new token within privacyIDEA, you can switch of the old RADIUS server.

For strategies on enrolling token see What are possible rollout strategies?.

Setup translation

The translation is performed using grunt. To setup the translation environment do:

npm update -g npm
# install grunt cli in system
sudo npm install -g grunt-cli

# install grunt in project directory
npm install grunt --save-dev
# Install grunt gettext plugin
npm install grunt-angular-gettext --save-dev

This will create a subdirectory node_modules.

To simply run the German translation do:

make translate

If you want to add a new language like Spanish do:

cd po
msginit -l es
cd ..
grunt nggettext_extract
msgmerge po/es.po po/template.pot > po/tmp.po; mv po/tmp.po po/es.po

Now you can start translating with your preferred tool:

poedit po/es.po

Finally you can add the translation to the javascript translation file privacyidea/static/components/translation/translations.js:

grunt nggettext_compile

Note

Please ask to add this translation to the Make directive translation or issue a pull request.

How can I setup HA (High Availability) with privacyIDEA?

privacyIDEA does not track any state internally. All information is kept in the database. Thus you can configure several privacyIDEA instances against one DBMS [1] and have the DBMS do the high availability.

Note

The passwords and OTP key material in the database is encrypted using the encKey. Thus it is possible to put the database onto a DBMS that is controlled by another database administrator in another department.

HA setups

When running HA you need to assure to configure the pi.cfg file on all privacyIDEA instances accordingly. You might need to adapt the SQLALCHEMY_DATABASE_URI accordingly.

Be sure to set the same SECRET_KEY and PI_PEPPER on all instances.

Then you need to provide the same encryption key (file encKey) and the same audit signing keys on all instances.

Using one central DBMS

If you already have a high available, redundant DBMS - like MariaDB Galera Cluster - which might even be addressable via one cluster IP address the configuration is fairly simple. In such a case you can configure the same SQLALCHEMY_DATABASE_URI on all instances.

Using MySQL master-master-replication

If you have no DBMS or might want to use a dedicated database server for privacyIDEA, you can setup one MySQL server per privacyIDEA instance and configure the MySQL servers to run in a master-master-replication.

Note

The master-master-replication only works with two MySQL servers.

There are some good howtos out there like [2].

Footnotes

[1]	Database management system

[2]	https://www.digitalocean.com/community/tutorials/how-to-set-up-mysql-master-master-replication.

MySQL database connect string

You can use the python package MySQL-python or PyMySQL.

PyMySQL is a pure python implementation. MySQL-python is a wrapper for a C implementation. I.e. when installing MySQL-python your python virtualenv, you also need to install packages like python-dev and libmysqlclient-dev.

Depending on whether you are using MySQL-python or PyMySQL you need to specify different connect strings in SQLALCHEMY_DATABASE_URI.

MySQL-python

connect string: mysql://u:p@host/db

Installation

Install a package libmysqlclient-dev from your distribution. The name may vary depending on which distritubtion you are running:

pip install MySQL-python

PyMySQL

connect string: pymysql://u:p@host/db

Installation

Install in your virtualenv:

pip install pymysql-sa
pip install PyMySQL

Are there shortcuts to use the Web UI?

I do not like using the mouse. Are there hotkeys or shortcuts to use the Web UI?

With version 2.6 we started to add hotkeys to certain functions. You can use ‘?’ to get a list of the available hotkeys in the current window.

E.g. you can use alt-e to go to the Enroll Token Dialog and alt-r to actually enroll the token.

For any further ideas about shortcuts/hotkeys please drop us a note at github or the google group.

How to copy a resolver definition?

Creating a user resolver can be a time consuming task. Especially an LDAP resolver needs many parameters to be entered. Sometimes you need to create a second resolver, that looks rather the same like the first resolver. So copying or duplicating this resolver would be great.

You can create a similar second resolver by editing the exiting resolver and entering a new resolver name. This will save this modified resolver definition under this new name. Thus you have a resolver with the old name and another one with the new name.

Cryptographic considerations of privacyIDEA

Encryption keys

The encryption key is a set of 3 256bit AES keys. Usually this key is located in a 96 byte long file “enckey” specified by PI_ENCFILE in The Config File. The encryption key can be encrypted with a password.

The three encryption keys are used to encrypt

• data like the OTP seeds and secret keys stored in the Token table,
• password of resolvers to connect to LDAP/AD or SQL (stored in the ResolverConfig table)
• and optional additional values.

OTP seeds and passwords are needed in clear text to calculate OTP values or to connect to user stores. So these values need to be stored in a decryptable way.

Token Hash Algorithms

OTP values according to HOTP and TOTP can be calculated using SHA1, SHA2-256 and SHA2-512.

PIN Hashing

Token PINs are managed by privacyIDEA as the first of the two factors. Each token has its own token PIN. The token PIN is hashed with a seed with SHA2-256 and stored in the Token database table.

This PIN hashing is performed in lib.crypto:hash.

Administrator Passwords

privacyIDEA can manage internal administrators using The pi-manage Script. Internal administrators are stored in the database table Admin.

The password is stored using a PBKDF with SHA512 with 10023 rounds. The hash is salted and peppered. While the salt is stored in the Admin table created randomly for each admin password the pepper is unique for one privacyIDEA installation and stored in the pi.cfg file.

This way a database administrator is not able to inject rogue password hashes.

The admin password hashing is performed in lib.crypto:hash_with_pepper.

Audit Signing

The audit log is digitally signed. (see Audit and The Config File).

The audit log can be handled by different modules. privacyIDEA comes with an SQL Audit Module.

For signing the audit log the SQL Audit Module uses the RSA keys specified with the values PI_AUDIT_KEY_PUBLIC and PI_AUDIT_KEY_PRIVATE in The Config File.

By default the installer generates 2048bit RSA keys.

The audit signing is performed in lib.crypto:Sign.sign using SHA2-256 as hash function.

Policies

How to disable policies?

I create an evil admin policy and locked myself out. How can I disable a policy?

You can use the pi-manage command line tool to list, enable and disable policies. See

pi-manage policy -h

How do policies work anyway?

Policies are just a set of definitions. These definitions are ment to modify the way privacyIDEA reacts on requests. Different policies have different scopes where they act.

admin policies define, what an administrator is allowed to do. These policies influence endpoints like /token, /realm and all other endpoints, which are used to configure the system. (see Admin policies)

user policies define, how the system reacts if a user is managing his own tokens. (see User Policies)

authentication and authorization policies influence the /validate/ endpoint (Validate endpoints).

The Authentication policies define if an authentication request would be successful at all. So it defines how to really check the authentication request. E.g. this is done by defining if the user has to add a specific OTP PIN or his LDAP password (see otppin).

The Authorization policies decide, if a user, who would authentication successfully is allowed to issue this request. I.e. a user may present the right credentials, but he is not allowed to login from a specific IP address or with a not secure token type (see tokentype).

How is this technically achieved?

At the beginning of a request the complete policy set is read from the database into a policy object, which is a singleton of PolicyClass (see Policy Module).

The logical part is performed by policy decorators. The decorators modify the behaviour of the above mentioned endpoints.

Each policy has its own decorator. The decorator can be used on different functions, methods, endpoints. The decorators are implemented in api/lib/prepolicy.py and api/lib/postpolicy.py.

PrePolicy decorators are executed at the beginning of a request, PostPolicy decoratros at the end of the request.

A policy decorator uses one of the methods get_action_value or get_policies.

get_policies is used to determine boolean actions like passonnotoken_policy.

get_action_value is used to get the defined value of non-boolean policies like otppin.

All policies can depend on IP address, user and time. So these values are taken into account by the decorator when determining the defined policy.

Note

Each decorator represents one policy and defines its own logic i.e. checking filtering for IP address and fetching the necessary policy sets from the policy object.

Performance considerations

You can test performace using the apache bench from the apache utils. Creating a simple pass token for a user, eases the performance testing.

Then you can run

ab -l -n 200 -c 8 -s 30 ‘https://localhost/validate/check?user=yourUser&pass=yourPassword‘

The performance depends on several aspects like the connection speed to your database and the connection speed to your user stores.

Processes

You should run several processes and threads. You might start with the number of processes equal to the number of your CPU cores. But you should evaluate, which is the best number of processes to get the highest performance.

Config caching

Starting with privacyIDEA 2.15 privacyIDEA uses a Cache per instance and process to cache system configuration, resolver, realm and policies.

As the configuration might have been changed in the database by another process or another instance, privacyIDEA compares a cache timestamp with the timestamp in the database. Thus at the beginning of the request privacyIDEA reads the timestamp from the database.

You can configure how often the timestamp should be read using the pi.cfg variable PI_CHECK_RELOAD_CONFIG. You can set this to seconds. If you use this config value to set values higher than 0, you will improve your performance. But: other processes or instances will learn later about configuration changes which might lead to unexpected behaviour.

Logging

Choose a logging level like WARNING or ERROR. Setting the logging level to INFO or DEBUG will produce much log output and lead to a decrease in performance.

Response

You can strip the authentication response, to get a slight increase in performace, using the policy no_details_on_success.

Clean configuration

Remove unused resolvers and policies. Have a realm with several resolvers is a bit slower than one realm with one resolver. Finding the user in the first resolver is faster than in the last resolver. Although e.g. the LDAP resolver utilizes caching.

Also see What happens in the tokenview?.

What happens in the tokenview?

A question which comes up often is why you can not view hundrets of tokens in the tokenview. Well - you are doing - you are just paging through the list ;-)

Ok, here it what happens in the tokenview.

The tokenview fetches a slice of the tokens from the token database. So, if you configure the tokenview to display 15 tokens, only 15 tokens will be fetched using the LIMIT and OFFSET mechanisms of SQL.

But what really influences the performance is the user resolver part. privacyIDEA does not store username, givenname or surname of the token owner. The token table only contains a “pointer” to the user object in the userstore. This pointer consists of the userresolver ID and the user ID in this resolver. This is usefull, since the username or the surname of the user may change. At least in Germany the givenname only changes in very rare cases.

This means that privacyIDEA needs to contact the userstore, to resolve the user ID to a username and a surname, givenname. Now you know that you will create 100 LDAP requests, if you choose to display 100 tokens on one page.

Although we are doing some LDAP caching, this will not help with new pages.

We very much recommend using the search capabilities of the tokenview.

How to mitigate brute force and lock tokens

For each failed authentication attempt privacyIDEA will increase a fail counter of a token. If the maximum allowed fail counter is reached, authentication with this token is not possible anymore. Starting with version 2.20 the administrator can define a timeout in minutes. The the last failed authentication is more than these specified minutes ago, a successful authentication will reset the fail counter and access will be granted. See Automatically clearing Failcounter.

The failcounter avoids brute force attacks which guess passwords or OTP values. Choose a failcounter clearing timeout, which is not too long. Otherwise brute force would also lock the token of the user forever.

Another possibility to mitigate brute force is to define an authorization policy with the action auth_max_fail. This will check, if there are too many failed authentication requests during the specified time period. If there are, even a successful authentication will fail. This technique uses the audit log, to search for failed authentication requests. See auth_max_fail.

Note

Some parts are marked as “(TODO) Not yet implemented”. These are components that have not been migrated from 1.5 to 2.0. If you are missing an important, not-yet-migrated part, drop us a note!

If you are missing any information or descriptions file an issue at github (which would be the preferred way), drop a note to info(@)privacyidea.org or go to the Google group.

This will help us a lot to improve documentation to your needs.

Thanks a lot!

Indices and tables

• Index
• Module Index
• Search Page