UNSOLVED
M
Mickeymonk
11 Posts
0
8027
October 14th, 2010 03:00
Scheduled savegrp will not auto restart after killing process id from command line
Hi,
we are running Networker 7.6 on a Solaris 10 server. I had to start a savegrp from the command but the savegrp ran for 2 days and was sitting idle on 1 client and was never going to complete so I had do a ps -ef | grep savegrp | grep , and then kill -9 the process id.
That worked fine but the problem is that in the Networker console the application still thinks that the group backup is running. The problem is that the daily schedule will no longer auto-start and reports the following message in the daemon.log. "42506 10/14/10 3:00:01 AM nsrd savegroup warning: Group GWP_PRD_FS_Daily aborted, savegroup is already running". I know that restarting NW will clear this issue but that's not the solution I'm looking for. We have several other groups running and would really like to resolve this without any disruptions to the application. Does anybody know a way around this?
DavidHampson-rY
1 Rookie
•
294 Posts
1322
0
Posted October 14th, 2010 05:00
How about before that, had you tried stopping from the console before issuing the kill command?
Have you tried starting the group from the command line and if so what happened?
DavidHampson-rY
1 Rookie
•
294 Posts
1322
0
Posted December 10th, 2012 08:00
Hi Zachschi
We haven't particularly got a solution to it; we have restarted Networker to resolve this and similar issues and once we have also cleared out the jobsdb before restarting as that appears to be the cause of the problem.
Mickeymonk
11 Posts
1322
0
Posted October 14th, 2010 05:00
Yes I tried to stop the group from the console but because I started the group from the command line I get the following error "Groups must be started either automatically or from the GUI and must be currently running in order to be stopped"
Mickeymonk
11 Posts
1322
0
Posted October 14th, 2010 05:00
Thanks for responding David. Yes I tries to stop from the GUI first and for that error "Groups must be started either automatically or from the GUI and must be currently running in order to be stopped". I then killed the associated savegrp process id. Even though the savegrp is no longer running (i.e. when I do a ps -ef now I do not see a prosess id for this group) the GUI has status still shows the savegrp as running. Whe I try to run the savegrp from the command line I get the following error "nsrd savegroup warning: Group GWP_PRD_FS_Daily aborted, savegroup is already running"
zacharyscheideg
8 Posts
1322
0
Posted November 13th, 2012 10:00
Did you ever get a way to fix the job? I am experiencing the same problem.
Anil_Kumar_Hari
13 Posts
1322
0
Posted November 14th, 2012 22:00
Hello,
Can you please try the command jobkill (which is similar to nsradmin prompt)? I am not totally sure if that exists in version 7.6
All,
We just had this issue on our Networker server, but after checking all of the nsr processes (e.g. ps -ef | grep -i nsr), we discovered that there were three 'nsrndmp_save' processes still running the 'dump' for those volumes, e.g.
root 25978 1396 0 17:52:07 ? 1:10 /usr/sbin/nsrndmp_save -T dump -F :ha -s nwserver -c nas1 -g srvr-nas1 root 21098 1396 0 Mar 04 ? 11:17 /usr/sbin/nsrndmp_save -T dump -F :mail2 -s nwserver -c nas2 -g srvr-nas2 root 21097 1396 0 Mar 04 ? 1:40 /usr/sbin/nsrndmp_save -T dump -F :project -s nwserver -c nas2 -g srvr-nas2 Killing each of those nsr dump processes (e.g. kill -9 25978) resolved this issue for us.
DavidHampson-rY
1 Rookie
•
294 Posts
1322
0
Posted October 14th, 2010 05:00
Did you try stopping the group from the console? If so what happened (or didn't happen)?