Issue
What were you trying to do that didn't work?
When systemd-239-45 is installed on the system and updated to systemd-239-78, systemd user units fail to stop with the following error message:
Feb 01 15:27:02 vm-rhel8 systemd[1]: Stopping User Manager for UID 0...
Feb 01 15:27:02 vm-rhel8 systemd[1258]: /usr/lib/systemd/user/systemd-exit.service:16: Failed to parse failure action specifier, ignoring: exit-force
Feb 01 15:27:02 vm-rhel8 systemd[1258]: systemd-exit.service: Service lacks both ExecStart= and ExecStop= setting. Refusing.
Feb 01 15:27:02 vm-rhel8 systemd[1258]: Failed to enqueue exit.target job: Unit systemd-exit.service has a bad unit file setting.
:
Feb 01 15:29:02 vm-rhel8 systemd[1]: user@0.service: State 'stop-sigterm' timed out. Killing.
Feb 01 15:29:02 vm-rhel8 systemd[1]: user@0.service: Killing process 1258 (systemd) with signal SIGKILL.
Feb 01 15:29:02 vm-rhel8 systemd[1]: user@0.service: Failed with result 'timeout'.
The reason for this is systemd user instances are not reexecuted upon update, but some of their unit files changed.
The following Upstream commit brings the functionality in systemd:
https://github.com/systemd/systemd/commit/463aef23a73d7850e0208cf4e96fd448693a755a
but this anyway requires using this new command in postinstall of the package.
I think this is necessary since even though the customers are supposed to reboot after updating, the systemd user instances should probably stop on such reboot.
Please provide the package NVR for which bug is seen:
systemd-239-78
How reproducible:
Always
Steps to reproduce
Downgrade to systemd-239-45.el8 and reboot
Login then update systemd to latest
Try stopping user@0.service
Expected results
The service stops
Actual results
Service doesn't stop and reports errors and remain in Stopping state
