re-occurring ID 702911 daemon.errors LMS 3.1 including MR-CALL_ON_NULL_OBJECT Ciscoworks LMS 3.1

[stuart hutchinson]

Hi,

Our monitoring system is picking up traps from our LMS3.1 Ciscoworks every few days , usually around 2.00am and some at 22:00. I've checked and these times are not when the archive runs and we don't appear to have any jobs running in the files in  /var/cron/spool/crontabs/ at this time.

The errors are r for  CI-E-ESRVTHEXCEPTION-Service thread closure for queue Discovery and MR-CALL_ON_NULL_OBJECT-Tried to make a call on a NULL repository object errors.

Are these real errors or can they be ignored?

The system is currently in failback mode . Also some log files are indicated on the LMS GUI that they are too large, but I've trimmed then down as much as possible.   Please see the summary of errors:

Apr 3 02:03:42 lprpncw002 DFM [ID 702911 daemon.error] [03-Apr-2014 02:03:42+326ms] t@50 Discovery #2

Apr 3 02:03:42 servername DFM [ID 702911 daemon.error] CI-E-ESRVTHEXCEPTION-Service thread closure for queue Discovery #2 threw

Apr 3 02:03:42 servername2 DFM [ID 702911 daemon.error] exception

Apr 3 02:03:42 servername DFM [ID 702911 daemon.error] MR-CALL_ON_NULL_OBJECT-Tried to make a call on a NULL repository object; in

Apr 3 02:03:42 servername DFM [ID 702911 daemon.error] file

Apr 3 02:03:42 servername DFM [ID 702911 daemon.error] "/work/greencurrent/DFM-3.1.3.0/8/smarts/auto-discovery/solaris2.8-sparc/o

Apr 3 02:03:42 servername DFM [ID 702911 daemon.error] ptimize/AD_Data.h" at line 8866

Apr 3 02:03:42 lprpncw002 DFM [ID 702911 daemon.error]

Thanks for any advice you may have...

[AFROJ AHMAD]

Hi ,

the messages you are getting is from DFM .

this is happening because of The inconsistencies referring to generally can arise during rediscoveries of
networks which have changed since the previous discovery.  Then any objects in the current topology which are no longer valid are deleted during the discovery process.  However, some concurrent process may still try to access these objects simultaneously (i.e. during or soon after the deletion).

so basically it if from DFM application ,"it is is looking for objects that does not exist in the database"

you can ignore them ,it is not impacting the LMS application ,however to get rid of it. you need to reinitialize the DFM databases  as below :

From DOS prompt:
-Stop the CiscoWorks processes:
net stop crmdmgtd (wait 5 minutes for the processes to stop)

-Clear DFM databases:
perl dbRestoreOrig.pl dsn=dfmInv dmprefix=INV npwd=cisco
perl dbRestoreOrig.pl dsn=dfmFh dmprefix=FH npwd=cisco
perl dbRestoreOrig.pl dsn=dfmEpm dmprefix=EPM npwd=cisco

-take the back up of file DFM.rps at

CSCOpx/objects/smarts/local/repos/icf/DFM.rps

and back it up to CSCOpx/objects/smarts/local/repos/icf/DFM.rps.orig  (copy the file and
rename it)

-Then, delete the file CSCOpx/objects/smarts/local/repos/icf/DFM.rps  file.

-Start the CiscoWorks Processes
net start crmdmgtd (wait 10 minutes for the processes to start again)

-When the services are up, check that the devices are discovered properly under DFM>
Device Management> Rediscover/Delete.

hope this information will help

Thanks-

Afroz

***Ratings Encourages Contributors ****

How to clear databases:
=================
WARNING: This process will delete the information about the devices, Events
and history in DFM. It will require a new discovery of the devices.

From DOS prompt:
-Stop the CiscoWorks processes:
net stop crmdmgtd (wait 5 minutes for the processes to stop)

-Clear DFM databases:
perl dbRestoreOrig.pl dsn=dfmInv dmprefix=INV npwd=cisco
perl dbRestoreOrig.pl dsn=dfmFh dmprefix=FH npwd=cisco
perl dbRestoreOrig.pl dsn=dfmEpm dmprefix=EPM npwd=cisco