-
Bug
-
Resolution: Unresolved
-
Normal
-
None
-
rhel-8.10
-
No
-
Low
-
rhel-systemd
-
ssg_core_services
-
5
-
False
-
False
-
-
None
-
None
-
None
-
None
-
Unspecified
-
Unspecified
-
Unspecified
-
None
What were you trying to do that didn't work?
A customer experiences "target is busy" when his RHEL8 system reboots.
Through enabling debug level, we can see that the first job systemd does is unmounting /run/user/XXX mounts, which ends up with "target is busy" since user@XXX.service} didn't stop yet (nor {{user-runtime-mount@XXX.service executed):
Mar 18 02:48:50.579146 ... systemd[1]: user-8743411.slice: Installed new job user-8743411.slice/stop as 5697067 Mar 18 02:48:50.580700 ... systemd[1]: run-user-8743411.mount: Installed new job run-user-8743411.mount/stop as 5697210 Mar 18 02:48:50.580706 ... systemd[1]: user@8743411.service: Installed new job user@8743411.service/stop as 5697069 Mar 18 02:48:50.580721 ... systemd[1]: user-runtime-dir@8743411.service: Installed new job user-runtime-dir@8743411.service/stop as 5697068 Mar 18 02:48:50.626184 ... systemd[1]: run-user-8743411.mount: About to execute: /usr/bin/umount /run/user/8743411 -c Mar 18 02:48:50.626713 ... systemd[1]: run-user-8743411.mount: Forked /usr/bin/umount as 677506 Mar 18 02:48:50.626796 ... systemd[1]: run-user-8743411.mount: Changed mounted -> unmounting Mar 18 02:48:50.626815 ... systemd[1]: Unmounting /run/user/8743411... Mar 18 02:48:50.631178 ... systemd[677506]: run-user-8743411.mount: Executing: /usr/bin/umount /run/user/8743411 -c Mar 18 02:48:50.638473 ... umount[677506]: umount: /run/user/8743411: target is busy. : Mar 18 02:48:55.657909 ... systemd[1]: Stopping User Manager for UID 8743411...
Many other mounts are similarly affected.
I see no ordering cycle or such thing that could explain the behavior.
What is the impact of this issue to you?
Unclean shutdown
Please provide the package NVR for which the bug is seen:
systemd-239-82.el8_10.3
How reproducible is this bug?:
Always on customer system, by issuing a reboot.
