Hi,

        I'm having a fairly frustrating issue with part of our netbackup
setup any help/suggestions much appreciated!

My setup is as follows..

Solaris 8 on all Master/Media/Clients
Netbackup 4.5
Single DLT8000
L9 ROBOT (DL8000)


The problem revolves around the single DLT8000, we are doing a full
backup of our Informix DB every night using the L9 Robot and currently
use the L9 for our Logical Logs (usually one log every 5 or 6 minutes),
which means logical logs get queued up a fair bit during the full
archive! I don't mind this but we decided to move the logical logs on a
separate tape so they could go on uninterrupted.

The problem I'm having is that netbackup seems to be unloading the
current tape in the single DLT and then requesting a completely
different tape even when there is plenty of space on the tape that's in
there. I've set ltid to start with -nsu and this helped some what but to
my amazement netbackup continued to request a different tape after doing
just one logical log on the current tape!

NB Log on the client..


14:39:24 Initiating backup
14:39:31 INF - Data socket = nbmaster02.1781
14:39:31 INF - Name socket = nbmaster02.1645
14:39:31 INF - Frozen image = 0
14:39:31 INF - Backup copy = 0
14:39:31 INF - Master server = nbmaster02
14:39:31 INF - New data socket = nbmaster02.1067
14:39:32 INF - Backup id = dbserver01-admin_1052833168
14:39:32 INF - Compression = 0
14:39:32 INF - Backup time = 1052833168
14:39:32 INF - Encrypt = 0
14:39:32 INF - Multiplexing = 0
14:39:32 INF - Client read timeout = 800
14:39:32 INF - Media mount timeout = 300

Right tape ->> 14:39:36 INF - Waiting for mount of media id LLG202 on
server nbmaster02.

14:40:00 INF - Beginning backup on server nbmaster02 of client
dbserver01-admin.
14:40:22 INF - Server status = 0
14:40:23 INF - Server status = 0
14:40:26 INF - Backup by informix on client dbserver01-admin using
policy InformixLogicalLogs: the requested operation was successfully
completed.


Ok this one is fine. It requested the correct tape AND used it, fine.

This is the next log..

14:40:26 Initiating backup
14:40:32 INF - Data socket = nbmaster02.3115
14:40:32 INF - Name socket = nbmaster02.4222
14:40:32 INF - Frozen image = 0
14:40:32 INF - Backup copy = 0
14:40:33 INF - Master server = nbmaster02
14:40:33 INF - New data socket = nbmaster02.1764
14:40:33 INF - Backup id = dbserver01-admin_1052833229
14:40:33 INF - Compression = 0
14:40:33 INF - Backup time = 1052833229
14:40:33 INF - Encrypt = 0
14:40:33 INF - Multiplexing = 0
14:40:33 INF - Client read timeout = 800
14:40:33 INF - Media mount timeout = 300

Wrong tape ->> 14:40:57 INF - Waiting for mount of media id LLG103 on
server nbmaster02.

14:41:18 INF - Server status = 150
14:41:19 INF - Server status = 150
14:41:22 INF - Backup by informix on client dbserver01-admin using
policy InformixLogicalLogs: termination requested by administrator.

Help! As you can see it trys to grab another tape from the pool and NOT
the tape already in there!

If I kill this job the loop begins again.. One backs up fine then the
second one asks for a different tape.. I also have allow multiple
retentions defined as when the tape was first assigned we used a
different retention period, would this cause a problem? 

It's got me completely baffled! Any help appreciated.

Cheers,

Samuel Stocker