dmgtd not stop the process

Answered Question
Jul 14th, 2008

/etc/init.d/dmgtd stop or star.not stop all the process or not start all the process. after those command the deamon manager not properly stop. or when use start. many of the process not start. the only way to back to normal I should use init 6 to start the whole system. any help!!! thanks

Correct Answer by Joe Clarke about 8 years 7 months ago

These steps were for RME 3.x, and have no bearing on config archive failing. They should have never been done under RME 4.0. The EssentialsDb* processes were part of LMS 2.2, and were named RMEDb* in LMS 2.5 and higher.

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 5 (1 ratings)
Joe Clarke Mon, 07/14/2008 - 09:03

What processes do not stop when running /etc/init.d/dmgtd stop? What version of Common Services do you have?

eliaspaulos Mon, 07/14/2008 - 12:48

simple way to explain even when I did stop and trying to strat. I got a message daemon manager alreday running. this is talking about after 10 -15 minutes waiting time to stop dmtgd. I don't know why takes so long? use to be 2-3 minutes.

CS- version is 3. 0 .6

process including ANI and all default process.

Joe Clarke Mon, 07/14/2008 - 12:53

The dmgtd shutdown can take a long time, but when the "dmgtd stop" command completes, all of the daemons should be down (unless you got an error indicating all the daemons could not be stopped).

Please post the output of ps -ef with dmgtd running. Also, post the output of ptree PID where PID is the PID of the dmgtd process. Then, run /etc/init.d/dmgtd stop. Then capture the ps -ef output again, and post it.

Just as a note, the dmgtd script should have the following cksum in CS 3.0.6:

757584272 23561 /etc/init.d/dmgtd

eliaspaulos Mon, 07/14/2008 - 13:33

I have attached the out put.I don't know how to check cksum for dmgtd script.

**responce I got when I start dmgtd stop.

/etc/init.d/dmgtd: /opt/dm_pids: cannot create


Responce I got after stop

WARNING: Daemon Manager terminated with SIGKILL.

WARNING: Please check if all processes have been terminated using

WARNING: the command - "ps -ef|grep CSCOpx" and

WARNING: terminate them if any processes are running.

out put for # ps -ef|grep CSCOpx

casuser 701 1 0 Jul 09 ? 0:06 /opt/CSCOpx/bin/cwjava -cw /opt/CSCOpx -cw:jre lib/jre -server -cp:p MDC/tomcat

casuser 727 1 0 Jul 09 ? 0:02 /opt/CSCOpx/bin/cwjava -cw:jre /opt/CSCOpx/MDC/jre -cp:p /opt/CSCOpx/MDC/tomcat

casuser 728 1 0 Jul 09 ? 0:01 /opt/CSCOpx/MDC/jre/bin/../bin/sparc/native_threads/java -cp /opt/CSCOpx/lib/cl

casuser 730 1 17 Jul 09 ? 296:27 /opt/CSCOpx/lib/jre/bin/java -server -Xmx1024m -Xminf0.1 -Xmaxf0.1 -XX:MaxPermS

casuser 8333 8332 0 Jul 13 ? 0:01 /opt/CSCOpx/objects/perl/bin/perl /opt/CSCOpx/bin/ /cw2000/Backup_Dir

casuser 8332 205 0 Jul 13 ? 0:00 sh -c /opt/CSCOpx/bin/perl /opt/CSCOpx/bin/ /cw2000/Backup_Dir /var/ad

root 11212 9988 0 14:35:57 pts/1 0:00 grep CSCOpx

casuser 8438 8437 2 Jul 13 ? 14:17 ls -lR /var/adm/CSCOpx/files/rme/dcma

casuser 8437 8333 0 Jul 13 ? 0:00 sh -c ls -lR /var/adm/CSCOpx/files/rme/dcma >> /cw2000/Backup_Dir/size

casuser 1025 1 0 Jul 09 ? 7:23 /opt/CSCOpx/bin/cwjava -cp:a lib/classpath/servlet.jar -Dvbroker.agent.port=423

casuser 1020 1 0 Jul 09 ? 0:00 /opt/CSCOpx/lib/vbroker/bin/osagent -p 42342

casuser 1029 1 0 Jul 09 ? 0:06 /opt/CSCOpx/bin/cwjava -cp:a /opt/CSCOpx:/opt/CSCOpx/lib/classpath:/opt/CSCOpx/

casuser 1080 1 0 Jul 09 ? 0:00 /opt/CSCOpx/objects/db/bin/diskWatcher diskWatcher

Joe Clarke Mon, 07/14/2008 - 14:02

It looks like something bad has happened with a backup that started running yesterday. It could very well be that your dcma directory is so huge that it's taking hours for the following command to complete:

ls -lR /var/adm/CSCOpx/files/rme/dcma

In that case, I would look at removing everything under the difffiles subdirectory as well as enabling more aggressive config file purging.

In the meantime, if you kill off the backup processes (8437 8438 8332 8333) as well as all of the remaining CSCOpx processes, you should be able to start, then cleanly stop dmgtd.

eliaspaulos Tue, 07/15/2008 - 08:18

J. thank you so much, it is on good progress, the only thing where is "difffiles " directory can be found to delete the subdirectory?

other quick question when dmgtd start,should I star any process manually or every necessary process started by dmgtd start command?

here is the examples what I see after the start

EDS-TR Never started 0

EssentialsDbEngine Administrator has shut down this server 0

EssentialsDbMonitor Failed to run 0

FDRewinder Never started 0

FHPurgeTask Never started 0

I'm really appreciated your expertise.

Joe Clarke Tue, 07/15/2008 - 08:32

The difffiles subdirectory is under /var/adm/CSCOpx/files/rme/dcma.

The EssentialsDb* processes shouldn't exist in LMS 2.5+. If they are there, this points to a bad migration/upgrade. You can remove these processes using the pdreg -u command, but if there were other problems with the upgrade, you may encounter other issues.

eliaspaulos Tue, 07/15/2008 - 11:25

I haveunregistered EssentialsDb*

the only upgrade I did is 1 1/2 month ago for

CS 3.06, CM 4.0.11 and RME 4.0.5. those upgrade cause of this issue?should I reverse something?

since the upgrade specially Config archive job failed almost by 90%. TAC Eng gave me the attached instruction but non of the command and the instruction working. I got error for each command. thanks again!!

Correct Answer
Joe Clarke Tue, 07/15/2008 - 12:39

These steps were for RME 3.x, and have no bearing on config archive failing. They should have never been done under RME 4.0. The EssentialsDb* processes were part of LMS 2.2, and were named RMEDb* in LMS 2.5 and higher.


This Discussion