Unsolved
This post is more than 5 years old
2 Posts
0
698
October 2nd, 2015 10:00
Index problems on client with multiple pools
We are running Networker 7.4.4 on an older Sun box running an older version of the O/S.
One of our client machines has multiple pools. Only the index from the earliest backup pool of the night for this client shows up in 'nwrecover'. To make it worse, only the index from the most recent backup shows up even though the browse policy has been set to 'Month'. All other clients with single pools generate and retain indexes as expected.
Any ideas?
No Events found!
bingo.1
2.4K Posts
0
October 3rd, 2015 03:00
Sorry - you seem to lack a bit of NW fundamental knowledge. This also includes the terminology.
"One of our client machines has multiple pools." This is incorrect as pools do not belong to any client. They sort (save stream) data that of course must come from one client in your backup scenario (your data zone). However, file index data is exclusively stored on the NW server. Consequently, if you open the NW User GUI on the NW server host, it will - by default - open the local client backups (the ones of the NW server itself) and of course you will only see the index backups and not the backups for the client himself.
If you want to browse the client data, you can either
- run the NW User GUI on the client himself or
- you must select the right source client on any other host where you start nwrecover.
zebulon3
2 Posts
0
October 8th, 2015 07:00
Sorry 'bingo', you're wrong.... I lack a LOT of NW fundamental knowledge. That's what happens when people retire and there's a hiring freeze on.... (sigh).
The main problem we have at the moment is that even though the Index Browse and Retention Policies have been set to '1 Month' for the client with a number of Pools associated with it, only the indexes for the most recent full or incremental backup is retained. If I go to /opt/nsr/index, where all the indexes are stored on the backup server, all the clients, except the one in question, have multiple index files in their 'db6' directory. The client in question only has the index file created that morning.
ble1
4 Operator
•
14.3K Posts
0
October 14th, 2015 00:00
In other words, your problem is that index files are retained only for last backup and if I allowed to guess, I would say that is until next nsrim check which will remove it. This has nothing to do with pools or retention, but rather misconfiguration on NSR_client or DNS/hosts. What usually happens is that name of client and the one used for backup are not the same (eg. cluster) and this could be seen with older code (which you have btw). I can't say what is wrong as this needs several checks to be performed, but you can try quick fix and place into backup command save -c - this will force index to be saved under name of NSR_CLIENT you use for backup.