03-19-2019 08:18 AM
Hi,
we are having failure on our db2 schedule backup. it was running fine all the while, but suddenly this morning the faiilure happen. on netbackup activity status shows below msg:
Mar 19, 2019 3:10:20 PM - Error bprd (pid=15413) Unable to write progress log </tmp/archive.log> on client <client> Policy=policy Sched=archive_infinite
Mar 19, 2019 3:10:22 PM - Error bprd (pid=15413) CLIENT <client> POLICY <policy> SCHED archive_infinite EXIT STATUS 130 (system error occurred)
there is no update or changes done on the netbackup nor the server itself. please seeking your kindness to help us with this matter.
03-19-2019 09:09 AM
Hello,
has /tmp on the client sufficient free space and permissions? Can you write to it when logged in interactively?
File system backup for this client is working?
Regards
Michal
03-22-2019 01:28 AM
hi Michal,
we already check the /tmp/archive.log space are sufficient. and also we already clear the archive.log file and rerun the backup, still failure.
Then we disable writing to archive.log command in the script and rerun the backup, seems like its working. so at the moment the backup running without writing the logs in archive.log.
is there anyway that we can have the backup running again with writing to the archive.log file?
03-22-2019 01:57 AM
Speak to server owner of db2 client to check permissions on /tmp.
The user account used for db2 backup should have full write access.
Default permissions of /tmp normally has 'sticky bit' set, i.e. 1777, which 'ls -l' shows as drwxrwxrwt.
Maybe someone has changed this?
03-24-2019 07:17 PM
Hi Marianne,
meaning the archive.log shud be 777 on permission in order to make the backup run? our current environment now :-
-rw-r--r-- 1 db2inst1 db2iadm1 archive.log....
Should i change it to 777?
03-25-2019 12:31 AM
I was referring to the /tmp folder permissions, not the log file.
If the file is indeed created, then the problem is not on the db2 client, but rather with comms between the master and the client.
You will probably need to dig into logs - the main ones are:
bprd on the master (NBU must be restarted after log folder is created)
bpdb2 and progress log on the client
(See the Troubleshooting chapter in the NBU for DB2 manual for full set of log folders).
If everything was working before and no change was made on NBU side, then you need to find out from all role players (network admins, server owners, dba, etc) if anything else has changed.
03-27-2019 06:31 AM
Found this technote https://www.veritas.com/support/en_US/article.100044915.html which might of interest.
03-27-2019 06:38 AM
Maybe you need to whitelist this path -> /tmp on client end?
https://www.veritas.com/support/en_US/article.100031821.html