From Fedora Project Wiki
 
Line 67: Line 67:
<!-- What other packages (RPMs) depend on this package?  Are there changes outside the developers' control on which completion of this feature depends?  In other words, completion of another feature owned by someone else and might cause you to not be able to finish on time or that you would need to coordinate?  Other upstream projects like the kernel (if this is not a kernel feature)? -->
<!-- What other packages (RPMs) depend on this package?  Are there changes outside the developers' control on which completion of this feature depends?  In other words, completion of another feature owned by someone else and might cause you to not be able to finish on time or that you would need to coordinate?  Other upstream projects like the kernel (if this is not a kernel feature)? -->


Systemd accepting the patch in a release that is in F18.
Systemd has accepted the patch in a release that is in F18.
 
== Contingency Plan ==
== Contingency Plan ==
<!-- If you cannot complete your feature by the final development freeze, what is the backup plan?  This might be as simple as "None necessary, revert to previous release behaviour."  Or it might not.  If you feature is not completed in time we want to assure others that other parts of Fedora will not be in jeopardy.  -->
<!-- If you cannot complete your feature by the final development freeze, what is the backup plan?  This might be as simple as "None necessary, revert to previous release behaviour."  Or it might not.  If you feature is not completed in time we want to assure others that other parts of Fedora will not be in jeopardy.  -->

Latest revision as of 12:47, 27 September 2012

SELinux Systemd Access Control

Summary

We need systemd to do SELinux access checking whether or not a process is allowed to manage a unit file.

Owner

  • Email: <dwalsh@redhat.com>

Current status

  • Targeted release: [Fedora 18]
  • Last updated: Sep 21 2012
  • Percentage of completion: 100%

selinux-policy fixes are in Fedora 18.

systemd-192-1.fc18.x86_64 selinux-policy-3.11.1-26.fc18.noarch


Detailed Description

In previous versions of Fedora/RHEL we were able to control which applications/users were able to start/stop services based on the label of the SYSV Init Script. With the advent of systemd we lost this ability since systemd starts and stops all services, and users and processes talk to systemd using systemctl.

Benefit to Fedora

We can better lock down services, for example: Currently NetworkManager needs to be able to start and stop the ntpd service. Currently the only way we can allow NetworkManager to start the ntpd service is to allow it to start/stop any service including security services like iptables/firewalld. Similarly with confined administrators, I would like to define a web administrator that can only manager files in /var/www/html, and start and stop the httpd service, without this feature I have to allow the confined admin to start/stop all services.

Scope

SELinux Policy needs to be written to govern which domains are able to control which units, (completed) Systemd has to be modified to look at the label of the calling process and the label of the unit file that the caller wants to manage, and then ask SELinux whether or not the caller is allowed the access.

How To Test

We can write a test suite to setup different success and failure situations. for example cause a process to run as NetworkManager_t, and execute the systemctl start ntpd.service (Should Succeed) Another attempt would be systemd start httpd.service (Should Fail) Try out confined administrators for webadm_t, and make sure he can only start and stop the httpd service.

Normal system testing should quickly show if there are bugs in the code.

Setup a confined user as guest_u and make sure he is not able to shut the machine down.

User Experience

Since unconfined_t is normal label of the user starting and stopping services, this fix should not effect normal administration of the system.

Dependencies

Systemd has accepted the patch in a release that is in F18.

Contingency Plan

We can continue allowing all access, but this is less secure then what we had in Fedora 15.

Documentation

This feature is to use SELinux to enforce Access Control to critical features like starting/stoping system services. Theoretically it should only effect the systemd developers and selinux policy writers.


Policy changes involve a new class called "service", with the following permissions


class service
{
       start
       stop
       status
       reload
       kill
       load
       enable
       disable
}

Since SELinux access checks and systemd access do not line up exactly, we have created a table to connect systemd (systemctl) commands with the appropriate SELinux access.

/*
  Define a mapping between the systemd method calls and the SELinux access to check.
  We define two tables, one for access checks on unit files, and one for
  access checks for the system in general.
  If we do not find a match in either table, then the "undefined" system
  check will be called.
*/
static const char *unit_methods[][2] = {{ "DisableUnitFiles", "disable" },
                                       { "EnableUnitFiles", "enable" },
                                       { "GetUnit", "status" },
                                       { "GetUnitByPID", "status" },
                                       { "GetUnitFileState",  "status" },
                                       { "Kill", "stop" },
                                       { "KillUnit", "stop" },
                                       { "LinkUnitFiles", "enable" },
                                       { "ListUnits", "status" },
                                       { "LoadUnit", "status" },
                                       { "MaskUnitFiles", "disable" },
                                       { "PresetUnitFiles", "enable" },
                                       { "ReenableUnitFiles", "enable" },
                                       { "Reexecute", "start" },
                                       { "Reload", "reload" },
                                       { "ReloadOrRestart", "start" },
                                       { "ReloadOrRestartUnit", "start" },
                                       { "ReloadOrTryRestart", "start" },
                                       { "ReloadOrTryRestartUnit", "start" },
                                       { "ReloadUnit", "reload" },
                                       { "ResetFailed", "stop" },
                                       { "ResetFailedUnit", "stop" },
                                       { "Restart", "start" },
                                       { "RestartUnit", "start" },
                                       { "Start", "start" },
                                       { "StartUnit", "start" },
                                       { "StartUnitReplace", "start" },
                                       { "Stop", "stop" },
                                       { "StopUnit", "stop" },
                                       { "TryRestart", "start" },
                                       { "TryRestartUnit", "start" },
                                       { "UnmaskUnitFiles", "enable" },
                                       { NULL, NULL }
};
static const char *system_methods[][2] = { { "ClearJobs", "reboot" },
                                          { "FlushDevices", "halt" },
                                          { "Get", "status" },
                                          { "GetAll", "status" },
                                          { "GetJob", "status" },
                                          { "GetSeat", "status" },
                                          { "GetSession", "status" },
                                          { "GetSessionByPID", "status" },
                                          { "GetUser", "status" },
                                          { "Halt", "halt" },
                                          { "Introspect", "status" },
                                          { "KExec", "reboot" },
                                          { "KillSession", "halt" },
                                          { "KillUser", "halt" },
                                          { "ListJobs", "status" },
                                          { "ListSeats", "status" },
                                          { "ListSessions", "status" },
                                          { "ListUsers", "status" },
                                          { "LockSession", "halt" },
                                          { "PowerOff", "halt" },
                                          { "Reboot", "reboot" },
                                          { "SetUserLinger", "halt" },
                                          { "TerminateSeat", "halt" },
                                          { "TerminateSession", "halt" },
                                          { "TerminateUser", "halt" },
                                          { NULL, NULL }
};

Policy for this would look like.

sesearch -A -s NetworkManager_t -c service
Found 5 semantic av rules:
  allow NetworkManager_t dnsmasq_unit_file_t : service { start stop status reload kill load } ; 
  allow NetworkManager_t nscd_unit_file_t : service { start stop status reload kill load } ; 
  allow NetworkManager_t ntpd_unit_file_t : service { start stop status reload kill load } ; 
  allow NetworkManager_t pppd_unit_file_t : service { start stop status reload kill load } ; 
  allow NetworkManager_t polipo_unit_file_t : service { start stop status reload kill load } ; 
sesearch -A -s webadm_t -c service
Found 1 semantic av rules:
  allow webadm_t httpd_unit_file_t : service { start stop status reload kill load enable disable } ;

Release Notes

  • systemd now verifies that calling process when it attempt to manage a service, using SELinux Policy

Comments and Discussion