Comment 3 for bug 1576987

Looking at the logs, this doesn't seem like a MAAS issue. MAAS monitors tgt daemon and starts it if it needs to be started or it attempts to restart it if it failed.

The following error log shows that MAAS found an error while querying for service status (meaning that something caused it to fail and MAAS doesn't know what).

апр 30 16:06:37 hostname maas.service_monitor[4530]: [ERROR] While monitoring service 'tgt' an error was encountered: Unable to parse the active state from systemd for service 'tgt', active state reported as 'deactivating'.

Then, looking at other logs this seems like the actual tgt service failed:

Setting up tgt (1:1.0.63-1ubuntu1) ...
Job for tgt.service failed because a fatal signal was delivered to the control process. See "systemctl status tgt.service" and "journalctl -xe" for details.
invoke-rc.d: initscript tgt, action "start" failed.

And the following:

апр 30 16:07:19 hostname systemd[1]: tgt.service: State 'stop-sigterm' timed out. Killing.
апр 30 16:07:19 hostname systemd[1]: Failed to start (i)SCSI target daemon.
апр 30 16:07:19 hostname systemd[1]: tgt.service: Failed with result 'signal'.

All in all, MAAS doesn't do anything other than try to start/stop tgt. If tgt is failing to start/install for whatever reason, then this is not MAAS' fault.

Can you please post the output of systemd to see what tgt faisl to start?