This could be related to many potential issues, but mainly these:
- If the system is mounting or serving storage via nfs, samba, cvfs, or other methods and those connections are stalled then it may take a long time to shutdown or never complete it. To confirm or discard this cause, umount all volumes and stop the snfs, smb, and cvfs services manually before attempting the shutdown.
- In some systems with Aja video boards, a known cause is the aja driver trying to unload too late (at the very end of the shutdown process). The sympthon is that most of the shutdown process seems to complete, but the aja driver is still loaded.
To solve it, create this extra daemon:
As root user, create this file:
With this content:
Description=Unload aja driver at shutdown, the aja daemon would do it too late
Now enable and actvate it:
systemctl enable sgo_ajashutdown.service
systemctl start sgo_ajashutdown.service