01-03-2014 05:57 AM
The last 3 weeks I have had the Netbackup 7.5.0.6 database just stop working.
Today This is the error I had when trying to log into see what was going on.
"Unable to obtail the server list from the Enterprise Meida Manager server database server is down (93)"
The Master/Medai server are the same server.I have 9GB of space which is low for this server I am guessing it is everything waiting to be backup and log files.
I have no idea where to start looking into this issue. Like I said this has been happening for about the last 3 weeks and nothing has really changed.
Solved! Go to Solution.
01-03-2014 06:49 AM
I suspect that the disk is filling up enough to cause NBU to shut down the DB to protect it from getting corrupted.
I think it does this somewhere arounf 90% full
M
01-03-2014 06:28 AM
Resolution:
If it is determined that nbpem is not running, it should be restarted. This can be accomplished from the Services interface on the Windows server.
After nbpem has restarted, start the EMM database from a command prompt, using the command nbdb_admin -start NBDB. (This command can be found in <install_path>\VERITAS\NetBackup\bin.)
Article URL http://www.symantec.com/docs/TECH61664
01-03-2014 06:49 AM
I suspect that the disk is filling up enough to cause NBU to shut down the DB to protect it from getting corrupted.
I think it does this somewhere arounf 90% full
M
01-03-2014 06:54 AM
Ah that could be the issue then. I will clear more space.
01-03-2014 06:55 AM
please see if you find any related info inthe server.log file
/usr/openv/db/log/server.log
please post the log if possible
01-03-2014 07:14 AM
Ok got up to 53GB free we will see what that does. Thank you. I will also try the restart command.
01-03-2014 07:29 AM
Sorry forgot to say this is a windows 2008 server R2.
I don't see a server.log any where.
01-03-2014 08:10 AM
it should be Installpath\Veritas\NetBackupDB\log\server.log if it is winddows.
01-03-2014 08:33 AM
Ah found it. Didn't look there. I will post if I find anything in the log.
01-03-2014 08:38 AM
Ok here is what I am seeing that looks Odd to me
01-03-2014 08:41 AM
Also seeing this from today
01-05-2014 02:25 PM
I hate to ask - but what can you apply 7.5.0.7? In that version parallelism (max_query_tasks) has been disabled by default. This prevents some very bad behavior similar to what you're sseeing.
See http://www.symantec.com/docs/TECH210414
** the above mentions the paralellism (max_query_tasks) issue in Item 3.
Otherthings I would suggest:
1. Rebuild your database.
from NetBackup\bin\ :
- Create a directory somewhere
- Run nbdb_backup -online /directory/you/created
*** this is just to make sure that we have a good backup.
- Shutdown everything except for the Sybase / dbsrv11 process.
- run nbdb_unload -rebuild -verbose
- When that's finished run: nbdb_admin -validate -full
** Hopefully that will look good.
If the problem persists:
2. In C:\Program Files\Veritas\NetBackupDB\conf edit the server.conf file (after saving a copy of course!) and increase the cache size parameters as follows
-c 500M
-ch (approximately 30% of physical memory)
*** Example - if your physical memory is 16G, set this to 4G : -ch 4G
-c 500M
Then stop/restart netbackup.
*** See: http://www.symantec.com/docs/HOWTO67149
If the problem persists:
3. In <install_path>\NetBackup\var\global add a file named emm.conf and put the following in it:
NUM_DB_CONNECTIONS=21
NUM_DB_BROWSE_CONNECTIONS=20
NUM_ORB_THREADS=31
And recycle.
Monitor the behavior after each change. I hope that 7.5.0.7 will fix the issue so you won't have to do anything else <fingers crossed>
Good luck
01-13-2014 08:42 AM
I will try the above and yes we will be update to .7 soon.
So far it seems to be a space issue, and I have been fighting to keep space there.
01-13-2014 11:24 AM
One other thing, note this:
01-22-2014 10:51 AM
Space was the issue. I have since fixed this issue. But have also rebuild the DB, and did a defrage, bo issues since than you.