We have detected some similar issues and a patch see PR1515 will improve even more the start of daemon, We have added local-fs.target to the After target (containing already remote-fs.target) to all daemons. Normally now systemd should wait until all local fs are up.
The failure setup has been also fixed in the meantime. On Monday, August 28, 2023 at 7:40:40 AM UTC+2 Silvio Schloeffel wrote: > Hi, > > not a showstopper but something that can make more work under certain > circumstances. > > Our RHEL8 based backup system has the filestorage mounted over nfs and > today we had scheduled a UPS shutdown/restart test. Everything worked as > requested, except the bareos service. > > The bareos-sd service was down. > [root@storage-serv ~]# systemctl status bareos-sd > ● bareos-sd.service - Bareos Storage Daemon service > Loaded: loaded (/usr/lib/systemd/system/bareos-sd.service; enabled; > vendor preset: disabled) > Active: failed (Result: exit-code) since Fri 2023-08-25 06:19:41 > UTC; 33min ago > Docs: man:bareos-sd(8) > Process: 1721 ExecStart=/usr/sbin/bareos-sd -f (code=exited, > status=1/FAILURE) > Main PID: 1721 (code=exited, status=1/FAILURE) > > Aug 25 06:19:41 storage-serv.musicdna.com systemd[1]: Started Bareos > Storage Daemon service. > Aug 25 06:19:41 storage-serv.musicdna.com bareos-sd[1721]: bareos-sd > ERROR TERMINATION > Aug 25 06:19:41 storage-serv.musicdna.com bareos-sd[1721]: > stored/stored_conf.cc:524 Unable to stat path > '/mnt/backup-storage/backupdata/storage' for device FileStorage1: > ERR=Datei oder Verzeichnis nicht gefunden > Aug 25 06:19:41 storage-serv.musicdna.com bareos-sd[1721]: Consider > setting Device Type if device is not available when daemon starts. > Aug 25 06:19:41 storage-serv.musicdna.com systemd[1]: bareos-sd.service: > Main process exited, code=exited, status=1/FAILURE > Aug 25 06:19:41 storage-serv.musicdna.com systemd[1]: bareos-sd.service: > Failed with result 'exit-code'. > > -> check if storage is mounted: > > 10.234.16.52:/backup 50T 6,1T 44T 13% /mnt/backup-storage > > Manual restart: > > [root@storage-serv ~]# systemctl start bareos-sd > [root@storage-serv ~]# systemctl status bareos-sd > ● bareos-sd.service - Bareos Storage Daemon service > Loaded: loaded (/usr/lib/systemd/system/bareos-sd.service; enabled; > vendor preset: disabled) > Active: active (running) since Fri 2023-08-25 06:56:24 UTC; 4s ago > Docs: man:bareos-sd(8) > Main PID: 3497 (bareos-sd) > Tasks: 2 (limit: 409922) > Memory: 1.4M > CGroup: /system.slice/bareos-sd.service > └─3497 /usr/sbin/bareos-sd -f > > > -> works. > > It looks like the bareos-sd daemon starts before all devices are > mounted. We mount the storage direct over the fstab, so it will mount as > early as possible but not before all network services are up and ready. > > So eventually it would be a solution to start the bareos-sd service > later -> eventually a delay. > > In your config you have: > > # Restart on failure, wait 30 seconds > Restart=on-failure > RestartSec=30 > > if this works, it could solve the problem. But it looks like it does not > work in my configuration. > > > Best > > Silvio > -- You received this message because you are subscribed to the Google Groups "bareos-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to bareos-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/bareos-users/1251c4a9-1fab-479a-a0e5-4f85d5c3b0e5n%40googlegroups.com.
