Linux-PAM (short for Pluggable Authentication Modules which evolved from the Unix-PAM architecture) is a powerful suite of shared libraries used to dynamically authenticate a user to applications (or services) in a Linux system.
It integrates multiple low-level authentication modules into a high-level API that provides dynamic authentication support for applications. This allows developers to write applications that require authentication, independently of the underlying authentication system.
Many modern Linux distributions support Linux-PAM (hereinafter referred to as “PAM”) by default. In this article, we will explain how to configure advanced PAM in Ubuntu and CentOS systems.
Before we proceed any further, note that:
- As a system administrator, the most important thing is to master how PAM configuration file(s) define the connection between applications (services) and the pluggable authentication modules (PAMs) that perform the actual authentication tasks. You don’t necessarily need to understand the internal working of PAM.
- PAM has the potential to seriously alter the security of your Linux system. Erroneous configuration can disable access to your system partially, or completely. For instance an accidental deletion of a configuration file(s) under /etc/pam.d/* and/or /etc/pam.conf can lock you out of your own system!
How to Check a Program is PAM-aware
To employ PAM, an application/program needs to be “PAM aware“; it needs to have been written and compiled specifically to use PAM. To find out if a program is “PAM-aware” or not, check if it has been compiled with the PAM library using the ldd command.
For example sshd:
$ sudo ldd /usr/sbin/sshd | grep libpam.so libpam.so.0 => /lib/x86_64-linux-gnu/libpam.so.0 (0x00007effddbe2000)
How to Configure PAM in Linux
The main configuration file for PAM is /etc/pam.conf and the /etc/pam.d/ directory contains the PAM configuration files for each PAM-aware application/services. PAM will ignore the file if the directory exists.
The syntax for the main configuration file is as follows. The file is made up of a list of rules written on a single line (you can extend rules using the
“\” escape character) and comments are preceded with
“#” marks and extend to the next end of line.
The format of each rule is a space separated collection of tokens (the first three are case-insensitive). We will explain the these tokens in subsequent sections.
service type control-flag module module-arguments
- service: actual application name.
- type: module type/context/interface.
- control-flag: indicates the behavior of the PAM-API should the module fail to succeed in its authentication task.
- module: the absolute filename or relative pathname of the PAM.
- module-arguments: space separated list of tokens for controlling module behavior.
The syntax of each file in /etc/pam.d/ is similar to that of the main file and is made up of lines of the following form:
type control-flag module module-arguments
This is a example of a rule definition (without module-arguments) found in the /etc/pam.d/sshd file, which disallows non-root logins when /etc/nologin exists:
account required pam_nologin.so
Understanding PAM Management Groups and Control-flags
PAM authentication tasks are separated into four independent management groups. These groups manage different aspects of a typical user’s request for a restricted service.
A module is associated to one these management group types:
- account: provide services for account verification: has the user’s password expired?; is this user permitted access to the requested service?.
- authentication: authenticate a user and set up user credentials.
- password: are responsible for updating user passwords and work together with authentication modules.
- session: manage actions performed at the beginning of a session and end of a session.
PAM loadable object files (the modules) are to be located in the following directory: /lib/security/ or /lib64/security depending on the architecture.
The supported control-flags are:
- requisite: failure instantly returns control to the application indicating the nature of the first module failure.
- required: all these modules are required to succeed for libpam to return success to the application.
- sufficient: given that all preceding modules have succeeded, the success of this module leads to an immediate and successful return to the application (failure of this module is ignored).
- optional: the success or failure of this module is generally not recorded.
In addition to the above are the keywords, there are two other valid control flags:
- include and substack: include all lines of given type from the configuration file specified as an argument to this control.
How to Restrict root Access to SSH Service Via PAM
As an example, we will configure how to use PAM to disable root user access to a system via SSH and login programs. Here, we want to disable root user access to a system, by restricting access to login and sshd services.
We can use the /lib/security/pam_listfile.so module which offers great flexibility in limiting the privileges of specific accounts. Open and edit the file for the target service in the /etc/pam.d/ directory as shown.
$ sudo vim /etc/pam.d/sshd OR $ sudo vim /etc/pam.d/login
Add this rule in both files.
auth required pam_listfile.so \ onerr=succeed item=user sense=deny file=/etc/ssh/deniedusers
Explaining the tokens in the above rule:
- auth: is the module type (or context).
- required: is a control-flag that means if the module is used, it must pass or the overall result will be fail, regardless of the status of other modules.
- pam_listfile.so: is a module which provides a way to deny or allow services based on an arbitrary file.
- onerr=succeed: module argument.
- item=user: module argument which specifies what is listed in the file and should be checked for.
- sense=deny: module argument which specifies action to take if found in file, if the item is NOT found in the file, then the opposite action is requested.
- file=/etc/ssh/deniedusers: module argument which specifies file containing one item per line.
Next, we need to create the file /etc/ssh/deniedusers and add the name root in it:
$ sudo vim /etc/ssh/deniedusers
Save the changes and close the file, then set the required permissions on it:
$ sudo chmod 600 /etc/ssh/deniedusers
From now on, the above rule will tell PAM to consult the /etc/ssh/deniedusers file and deny access to the SSH and login services for any listed user.
How to Configuring Advanced PAM in Linux
To write more complex PAM rules, you can use valid control-flags in the following form:
type [value1=action1 value2=action2 …] module module-arguments
Where valueN corresponds to the return code from the function invoked in the module for which the line is defined. You can find supported values from the on-line PAM Administrator’s Guide. A special value is default, which implies all valueN’s not mentioned explicitly.
The actionN can take one of the following forms:
- ignore: if this action is used with a stack of modules, the module’s return status will not contribute to the return code the application obtains.
- bad: indicates that the return code should be thought of as indicative of the module failing. If this module is the first in the stack to fail, its status value will be used for that of the whole stack.
- die: equivalent to bad but may terminate the module stack and PAM immediately returning to the application.
- ok: this instructs PAM that the system administrator thinks this return code should contribute directly to the return code of the full stack of modules.
- done: equivalent to ok but may terminate the module stack and PAM immediately returning to the application.
- N (an unsigned integer): equivalent to ok but may jump over the next N modules in the stack.
- Reset: this action clears all memory of the state of the module stack and restart with the next stacked module.
Each of the four keywords: required; requisite; sufficient; and optional, have an equivalent expression in terms of the
[...] syntax, which allow you to write more complicated rules and they are:
- required: [success=ok new_authtok_reqd=ok ignore=ignore default=bad]
- requisite: [success=ok new_authtok_reqd=ok ignore=ignore default=die]
- sufficient: [success=done new_authtok_reqd=done default=ignore]
- optional: [success=ok new_authtok_reqd=ok default=ignore]
The following is an example from a modern CentOS 7 system. Let’s consider these rules from the /etc/pam.d/postlogin PAM file:
#%PAM-1.0 # This file is auto-generated. # User changes will be destroyed the next time authconfig is run. session [success=1 default=ignore] pam_succeed_if.so service !~ gdm* service !~ su* quiet session [default=1] pam_lastlog.so nowtmp showfailed session optional pam_lastlog.so silent noupdate showfailed
Here is another example configuration from the /etc/pam.d/smartcard-auth PAM file:
#%PAM-1.0 # This file is auto-generated. # User changes will be destroyed the next time authconfig is run. auth required pam_env.so auth [success=done ignore=ignore default=die] pam_pkcs11.so nodebug wait_for_card auth required pam_deny.so account required pam_unix.so account sufficient pam_localuser.so account sufficient pam_succeed_if.so uid < 1000 quiet account required pam_permit.so password required pam_pkcs11.so session optional pam_keyinit.so revoke session required pam_limits.so -session optional pam_systemd.so session [success=1 default=ignore] pam_succeed_if.so service in crond quiet use_uid session required pam_unix.so
For more information, see the pam.d man page:
$ man pam.d
Lastly, a comprehensive description of the Configuration file syntax and all PAM modules can be found in the documentation for Linux-PAM.
PAM is a powerful high-level API that allows programs that rely on authentication to authentic users to applications in a Linux system. It’s powerful but very challenging to understand and use.
In this article, we’ve explained how to configure advanced features of PAM in Ubuntu and CentOS. If you have any questions or comments to share, use the feedback form below.
9 thoughts on “How to Configure and Use PAM in Linux”
Hello, I’m reading this article. And When reading the additional control flag, there be a repeat about include and substack:
include: include all lines of given type from the configuration file specified as an argument to this control.
substack: include all lines of a given type from the configuration file specified as an argument to this control.
As above, The Explain of include and substack is the same. I think It’s just a little mistake
Thanks, corrected the meaning of both in the article…
I am using the PAM module in my project. While using the pam_limits library I found that when we set both maxlogins and maxsyslogins in /etc/security/limits.conf, Functionality whichever is declared first in /etc/security/limits.conf that functionality is not working.
By looking into the pam_limit code what I found is that in pam_limits module functionality to support both configurations (maxlogins and maxsyslogins) at a time is not handled properly. Is it the really case ?? Or Is there something that I am missing. Any help is appreciated.
I am getting an error on executing this following command on my Ubuntu 18.04 version.
Please help me out
Please could you share the error message you getting?
I am using CentOS 8 on VMware and I want to set login time limits on my system’s local users using PAM. How can I do this I have done all the possible way. I don’t know where I am making mistake. Please help me to show the best way to do that.
We may have to work on a complete guide for this.
One of the strengths of Linux is how configurable it can be. One of the weaknesses is how complex it can be, especially for newbies without exposure to comparable concepts in Windows.
Are there any simple GUIs that script configuration say by asking the user to check boxes? I am thinking of things like the GUIs that exist for firewalls to make them simpler. An example would be a screen that allows you to check which users should have access to Firefox, which users should have remote access, which users should be able to open a terminal…
Thanks for your interesting article!
There are GUI tools for configuring various features of Linux, especially on modern Linux distributions. However, using the command line gives system administrators more power in configuring a Linux system. Thanks for sharing your thoughts with us. We are grateful.