Legato Networker - rebuilding indexes

[brewinp]

Am currently using Networker v5.5, 5.5.1 and 6.0.1 across 23 clients, running linux/unix. Recently I've noticed the clients are taking 3 hours to backup on a daily basis, but a further 12 hours is being spent each day by the server cross-checking indexes of all the clients. There are some days when the entire back-up process takes 3 days to complete.
My question is; does this have to do with the indexes being fragmented or too large? Also, how do you rebuild the indexes (assuming this would reduce the about of fragmentation)?
Thanks

 

[wallace88]

12 hours to cross check indexes seem high, but it also depends on the size of the indexes too.

Try checking your client indexes on the backup server with:

NetWorker V5.5.x: nsrck -F <client name>

NetWorker V6.x: nsrck -L6 <client name>

If you are running NetWorker V5.x on the backup server, you may need to run nsrck -F <client name> more than once. If it reports an error, then run it again. This is because when nsrck encounters a problem, it tries to correct it and then stops processing. In V6.x, it keeps checking all it is suppose to check.

Checking indexes can take a long time, so I would suggest doing them one at a time, starting with the backup server itself, followed by the clients that seem to be taking a long time to back up.

Idealy, your client should all be the same version as the backup server. Considering how trivial it is to upgrade the clients, there's no reason why not.

Note that NetWorker v5.x is at end of support life at the end of the year 2002.