01-19-2012 09:46 AM
Hello:
We have BI environment with db2 cluster, 1 admin host and 3 data hosts.
When I ran db2 restore command with load option for vendor library '/usr/openv/netbackup/bin/nbdb2.sl64', it was able to retrieve db2 backup images for admin, data 1-2 hosts, but it can't find the backup images for data host 3.
11:42:44.710 [12976632] <4> sqluvint: INF --- RETURN STATUS
11:42:44.710 [12976632] <4> sqluvint: return_code=[516]
<font color=red>11:42:44.710 [12976632] <4> sqluvint: description=[NetBackup Find Image Failed]</font>
11:42:44.710 [12976632] <4> sqluvint: sqluv_status=[11]
11:42:44.710 [12976632] <4> sqluvint: EXIT CALL[0.1] 00:00:00
<font color=red>11:42:44.710 [12976632] <4> sqluvint: END - DATABASE DQA02CDW NODE=19 SESSION=0 TIME=20120118</font>
However, bplist from data host 3 shows that the backup images for data host 3 exist:
$ /usr/openv/netbackup/bin/bplist -B -C GA016D28A -t 18 -l -R -s '01/18/2012' -e '01/19/2012' / |grep 20120118
-rw-rw---- uzdbedw2 bcuigrp 67129344 Jan 18 09:16 /DB2/DQA02CDW/node0023/20120118091644/DQA02CDW.0.uzdbedw2.node0023.1.20120118091644.1
-rw-rw---- uzdbedw2 bcuigrp 67129344 Jan 18 09:16 /DB2/DQA02CDW/node0020/20120118091644/DQA02CDW.0.uzdbedw2.node0020.1.20120118091644.1
-rw-rw---- uzdbedw2 bcuigrp 67129344 Jan 18 09:16 /DB2/DQA02CDW/node0022/20120118091644/DQA02CDW.0.uzdbedw2.node0022.1.20120118091644.1
-rw-rw---- uzdbedw2 bcuigrp 67129344 Jan 18 09:16 /DB2/DQA02CDW/node0017/20120118091644/DQA02CDW.0.uzdbedw2.node0017.1.20120118091644.1
-rw-rw---- uzdbedw2 bcuigrp 67129344 Jan 18 09:16 /DB2/DQA02CDW/node0018/20120118091644/DQA02CDW.0.uzdbedw2.node0018.1.20120118091644.1
-rw-rw---- uzdbedw2 bcuigrp 67129344 Jan 18 09:16 /DB2/DQA02CDW/node0024/20120118091644/DQA02CDW.0.uzdbedw2.node0024.1.20120118091644.1
<font color=red>-rw-rw---- uzdbedw2 bcuigrp 67129344 Jan 18 09:16 /DB2/DQA02CDW/node0019/20120118091644/DQA02CDW.0.uzdbedw2.node0019.1.20120118091644.1</font>
-rw-rw---- uzdbedw2 bcuigrp 67129344 Jan 18 09:16 /DB2/DQA02CDW/node0021/20120118091644/DQA02CDW.0.uzdbedw2.node0021.1.20120118091644.1
Please note that db2.conf for netbackup is on a shared path for all hosts and so it is the same. And vendor library file '/usr/openv/netbackup/bin/nbdb2.sl64' is also the same in bytes and timestamp. We even tried to use the file from data host 2 on data host 3 to see if it makes any difference at all, but got the same error.
Anyone has any clue, please let me know.
We have also opened a case with symantec but I just want to see if anyone has encountered similar issues.
Thanks.
Sheila
Solved! Go to Solution.
02-15-2012 08:57 AM
It turns out that the client name is mixed case in one of the bp config files. Once the client name is corrected to all uppcase, it is able to locate the backup images.
Thanks all,
Sheila
01-19-2012 12:21 PM
Are you trying the restore as the same user that performed the backup (uzdbedw2)?
Please ensure that you have bprd log folder on the master (Symantec Support will also need this log). If you need to create the folder, remember to restart NBU to activate the log.
01-19-2012 12:44 PM
Yes, Marianne, it is the same user id that did the backup and restore and the backup images on netbackup server are owned by the same user uzdbedw2.
Thanks.
Sheila
01-19-2012 12:55 PM
We need bprd log on master to see what's wrong...
01-20-2012 06:08 AM
Marianne:
I think that the storage group is handling the log as I don't have access to that box yet. I'm just a dba.
Thanks.
Sheila
02-15-2012 08:57 AM
It turns out that the client name is mixed case in one of the bp config files. Once the client name is corrected to all uppcase, it is able to locate the backup images.
Thanks all,
Sheila