BAckup stops but does not get aborted

Question

We have a Solaris client using NW 7.2.2 in which one folder around 47 GB does not get backed up completely.

We have reduced the Group inactivity timeout to 0 and run using savegrp. We have also tested it by putting it in a Test group alone.

It backs up few GB and then aborts. When we ran the group using savegrp, it backed up all 47 GB & then continued without any errors. It neveer wrote the index till we aborted.

After aborting, we got the error:

Aborting invalid connection from 0.0.0.0/21236 to 0.0.0.0/0., Connection timed out

Please suggest.

ble1 · Answer

Go to client and kill save command. I have seen situations like these when you have dodgy network connection between client and server and while you can kill the group save session is not aborted, but it is left open and keeps writing very slowly or not at all (locking the device).

amediratta · Answer

True but in that case the group would end.

The problem is that backups do not happen properly i.e. sometimes this situation may come up only after 10 GB or 20 GB etc.

No indexes get written even if all 47 GB gets backed up.

Main issue is proper backups.

ble1 · Answer

Another thing to check is if you have stale mnt point on that box - that may freeze your backup session and keep group running endlessly. If you do df -k as root on the client you will know if you have that or not (df won't return cmd prompt if there is stale mnt point). In that case killing stale mnt point will result in session to finish and group complete.

NetWorker

BAckup stops but does not get aborted

Was this post helpful?