my backups are timing out

[billhause]

I have a legato version 6. I'm backing up 7 clients. The data runs from less than 1GB of data to 100 GB of data. I force the system to do a full back up. 6 of the clients run fine. they are all small. The one that is 96GB get about 17GB of data backed up then goes inactive for 10 minutes and times out. Any help would be great.

Bill Hause

 

[wallace88]

Inactive means that the client stopped sending data after (in your case) 10 minutes. The unanswered question is why?

Check the back server's daemon.log to see if there are any clues during the backup period.

What type of file system are you backing up? Is the file system:

- on the client?
- on a network storage device (eg, Netapps)?
- RAID device?
- NFS remote mounted?

Does a manual backup succeed, or does it fail after 17GB?

What are you backing up, ie what type of data is it?

What o/s platforms? What version of NetWorker?

Is there a firewall between the client and the backup server?

Are there multiple NIC cards? Was there network congestion?

What's the server's parallelism? Group parallelism? Client parallelism?

The list of questions and possibilities are endless. Hopefully these questions can point ou in the right direction.

A workaround may be for you to set the group's inactivity timeout value to zero, thus telling NetWorker not to timeout. If backups work, then you either accept this workaround as a solution, or continue digging to determine what is causing the inactivity period.

 

[wallace88]

In addition, search through the articles in this forum for further hints. This issue has been duscussed many times. One recent thread was titled: "aborted due to inactivity"

 

[pmidds]

I think wallace88 has covered all bases but I'll let you know one of my experiences which sounds familiar. We had a file server of approximately 60 Gb and it only backed up about 35 % of this on a full dump.

On doing a manual backup the same thing happened. Looking at the logs I saw that allthough the client was set to backup saveset 'All' it was only backing up 2 of the 3 volumes ( failing on the data volume !!!! ) There were no errors reportd in the log and the backup was always reported as succesfull. There were over 50 clients in the group and it was only this one which had this problem.


If I did a manual backup of the data partition only this would work but not as part of a full server backup. The way I worked around it was to create two clients one to backup the first two partitions and the 2nd to backup the 3rd ( data ) partition. This worked fine.

The machine in question is no longer of this world and I never got to the bottom of why this was happening. However apart from having to create an extra client the workaround was painless.

 

[nandfred]

It's been my experience that when large file systems fails like that, it's due to the fact that either NIC's or switches are set to auto negociate instead of being locked at 100/MB full or half.