Networker refreshes to new daemon.raw only when backup server services are restarted. If not, the file can grow huge. We had a similar problem recently and upgrading to version 7.6 helped us. Daemon.raw has not grown that big very fast after that.

Hi,

Have a look at this Knowledgebase Article:

How to control the  size of the daemon log file

http://solutions.emc.com/EMCSolutionView.asp?id=esg53730&usertype=C

It explains what you need to do to limit the size that the daemon.raw grows to.  You will need to restart Networker for this change to take effect.

-Bobby

you don't have any savegroups running which have verbose enabled? Experienced that recently with one savegroup with multiple clients in it that included filesystem backups. Every single file in the backup was logged thus flooding the daemon.raw. A couple of Gigs per night where logged for millions of files...

One forum folk told me that there is an RFE opened for limit the log size in runtime, at the moment you would need to reboot NetWorker services and the daemon.raw would be regenerated if the configured limit size is reached.

Regards.

Hi!

I´ll open this old topic again because I still experience this problem from time to time.

According to this article:https://solutions.emc.com/emcsolutionview.asp?id=esg93582 , you should upgrade to 7.4.1.

I´m currently running '7.6.2 build 631' but still get the same issue.

I read through "How to control the size of the daemon log file"

http://solutions.emc.com/EMCSolutionView.asp?id=esg53730&usertype=C

There it says:

The daemon.log file size is only checked when NetWorker is started.

Therefore the maximum log file size should only be considered as a guideline and not as a hard rule.

The daemon.log will exceed this limit if NetWorker is not restarted for an extended period of time.

As I see this.. it doesn´t matter what limit you set if you don´t freekvently restart the Networker services?

Hello Henrik,

yes you are correct, if the networker services are not restarted the log will not be rotated.

As per the previous post from BGuevara, there is a Request for Enhancement open for this behaviour to be changed but as of the latest release 7.6.2 this behaviour is unchanged.

Regards,

Conor.

Does this rotation really works for daemon.raw?  I know I tested it in the past, but it didn't work for daemon.raw.  Last weekend I did rotation manually as my daemon.raw was over 100GB   On UNIX it is easy at least (cp /nsr/logs/daemon.log /nsr/logs/old/daemon.$TIMESTAMP; > /nsr/logs/daemon.log).

Hi Hrvoje,

Yes this works fine with daemon.raw also, confirmed with 7.6.1.2 on solaris 10.

pargs -e 13018

13018:  /usr/sbin/nsrd

........

envp[11]: NSR_MAXLOGSIZE=1048576

envp[12]: NSR_MAXLOGVERS=5

.....

# du -sh ./daemon.raw
2.1M   ./daemon.raw

# nsr_shutdown

# du -sh ./daemon.raw
2.6M   ./daemon.raw

# /etc/init.d/networker start

# du -sh ./daemon.raw
   1K   ./daemon.raw

Regards,

Conor.l

EMC NetWorker Release 7.6 Service Pack 1 Administration Guide [Page  468]

Note: The trimming mechanism functions only when you start nsrd on UNIX or the
NetWorker Backup and Recover Server service on Windows. The daemon and service do not
check periodically to determine if the log file has exceeded NSR_MAXLOGSIZE. If the nsrd
daemon or NetWorker Backup and Recover Server service runs for a long time, the log file can
still grow very large.

EMC® NetWorker® Release 7.6 Service Pack 1 Administration Guide P/N 300-011-327 REV A01

Thanks guys!

Does anyone know where you can follow the RFE for this issue (if it´s possible)?

Thanks.  I'll give it a try over the weekend.  I have updated version since I tried this last time so it might work now.