Bug 1197191

Summary:	Failed unmounting /etc. after reboot or shutdown
Product:	[openSUSE] PUBLIC SUSE Linux Enterprise Micro 5.2	Reporter:	Jose Lausuch <jalausuch>
Component:	Base	Assignee:	Jiri Srain <jsrain>
Status:	NEW ---	QA Contact:	Jiri Srain <jsrain>
Severity:	Normal
Priority:	P5 - None
Version:	Beta
Target Milestone:	---
Hardware:	Other
OS:	Other
Whiteboard:
Found By:	---	Services Priority:
Business Priority:		Blocker:	---
Marketing QA Status:	---	IT Deployment:	---
Attachments:	journal

Description Jose Lausuch 2022-03-16 17:39:21 UTC

Created attachment 857106 [details]
journal

After issuing a reboot command via `reboot`, `shutdown` or using `rebootmgrctl`,
the following message is shown in the prompt:

> [FAILED] Failed unmounting /etc

This is an extract of journal:

> Mar 15 15:30:29.995933 localhost systemd[1]: systemd-udevd.service: Succeeded.
> Mar 15 15:30:29.996791 localhost systemd[1]: Stopped Rule-based Manager for Device Events and Files.
> Mar 15 15:30:30.000997 localhost systemd[1]: Unmounting /etc...
> Mar 15 15:30:30.012292 localhost systemd[1]: systemd-update-utmp.service: Succeeded.
> Mar 15 15:30:30.012897 localhost systemd[1]: Stopped Update UTMP about System Boot/Shutdown.
> Mar 15 15:30:30.014790 localhost systemd[1]: Stopping Security Auditing Service...
> Mar 15 15:30:30.019518 localhost umount[5660]: umount: /etc: target is busy.
> Mar 15 15:30:30.021038 localhost systemd[1]: etc.mount: Mount process exited, code=exited, status=32/n/a
> Mar 15 15:30:30.021060 localhost systemd[1]: Failed unmounting /etc.
> Mar 15 15:30:30.032632 localhost auditd[1022]: The audit daemon is exiting.

The attached full journal log belongs to a system running on RPi 4, but this error also appears in the journal of openQA jobs, 
e.g. https://openqa.suse.de/tests/8329169#step/journal_check/20

This bug is probably not new, I just noticed about it while testing on RPi.
I don't think it's critical, but users might be confused by this message shown on the prompt.

Comment 1 Thorsten Kukuk 2022-03-17 12:57:43 UTC

That "bug" is normal and on a "normal" SLES you normally do not notice it with the standard filesystem layout, since the root filesystem is just mounted read-only.
You have processes which do the shutdown, and everything which is in use by this processes cannot be unmounted.