I am regularly having problems with my scheduled backups failing with the error message below
There are currently 3 backups running, and the system-wide limit is 3
If I do nothing about it then every backup fails every day. When I run "ps -ael" I find 3 processes of backup.pl and a "killall backup.pl" gets the backups running again.
This seems to show that the problems is due to backup.pl (presumably the scheduled backup task) running but not closing itself properly.
My backup schedule is as follows. Daily backup to dedicated server in same datacenter Daily backup to private server with dynamic IP on a sometimes temperamental internet connection in Thailand. Weekly backups of the same.
My thought is that it is probably the backup to our office server that is causing the problems, failing to establish a connection and hanging the process. I could disable this but I do like the comfort of an effort free offsite backup.
Does anyone have any suggestions?
All 3 servers are running virtualmin GPL on ubuntu (2 x 8.04 and 1 x 10.04)
Im getting the same error is there a resolution?
Hmm, I don't know if that is the way you wish to go about this, but there are configuration options about how many backups are allowed to run in parallel, and how long to wait for one to finish if that number is exceeded.
Those settings are under System Settings / Virtualmin Config / Backup and restore.
If a backup process hangs though that might not help, since every amount of parallel backups will be exceeded then at some point, and no amount of waiting helps. Maybe you might try to figure out if there really is a hanging backup process and why/where it hangs, if that is possible.
If it does hang, find out if it hangs always. If not, maybe some kind of watchdog script might help, which cron-runs at a time when all backups should be long finished, and which kills all remaining backup processes?
I've just updated to the newest version of virtualmin and I experience the same issue: The backup is done properly but the backup.pl script doesn't terminate after backup. Also the lockfile remains in the target backup folder (name_of_host.tar.gz.info.lock).
Is there a way to debug the backup process to see where exactly the script hangs?