Failed savegroup notification but backup still running

[morgaann]

Don't know if anyone has seen this before but I am running Networker 6.1.2 on a Solaris server and there is a savegroup running for an NT client. It is still backing up but I received the savegroup notification which states the following:

etormfs01:\data\users save: using `D:\Data\Users' for `D:\data\users'
* etormfs01:\data\users has been inactive for 30 minutes since Fri Oct 17 08:17:51 2003.
* etormfs01:\data\users is being abandoned by nsrexec.

I look at the Group Control status for that group and it says its finished.

Any ideas as to why it's still backing up when all processes point to the Group as having failed its backup.

Any help would be appreciated.

Thanks.

 

[sdeb]

This error happens when you backup something that takes a very long time... (such as thousands of small files) or when the Network connection has really poor throutputs...

You could bypass it by increasing the savegroup timeout (on the the group ressource) or even disabling it (set it to 0).

 

[sdeb]

It also happen sometimes that the backup session is not properly killed... an easy answer would be "Upgrade to 6.1.3" its a really stable version.

 

[morgaann]

Thanks for that info. Just one more thing. The fact that from the GUI it still shows the backup running does it mean it really is writing to tape or can I assume it's dead and attempt to restart.

Also I noted that the throughput when backing up is on average 300kb/s (very poor) but interestingly I backup the same client to another Networker server at the same time and the average throuhgput is 3Mb/s. How is this possible??

Thanks for your help.

 

[sdeb]

It must not be writing to tape but you'd better try to stop/restart services on that Networker client.
Because the backup session has not been cleanly closed.

If you backup at the same time, one of the both must take the network bandwith (I don't think so) or overload the client.
If you backup same disk / same files at different times with different servers and you have different throuputs (with same backup devices), there can be a network issue (backup flow going through different routes for example).

 

[sdeb]

Or if this client has multiple NIC's. One well configured and not the other one...

(try to avoid autonegociation as much as possible ... it could decrease the throughtput by 10 sometimes).

 

[morgaann]

Thanks again for the info. I will do some research and get this resolved.

Much appreciated.