LMS2.6 - CS & RME Errors

[Brendan O'Flynn]

Hello Folks,

When I tried to establish the SyslogAnalyzer Collection, I received error message SCS.png attached.

After I verified that SyslogAnalyzer was not running with message of either "Never started" or "Administrator has stopped this service", I tried restarting it and received error message RMEMon.png attached. This was also received when I noticed that processes RMEDbEngine and RMEDbMonitor are also in the same state (not started or won't start) and tried starting those processes.

At this stage I saw that there are "DCA Server is down or inaccessible" and JRM server is down or inaccessible" alerts on Common Service Home - see CS-Home.png. The process list also has the following processes as "Administrator has shut down this server": CTMJrmServer and jrm - these can be started OK. I'm also attaching dmgtd, pdshow and dcr debug logfiles.

Could anybody shed some light on this issue?

Kind Regards,

Brendan

[Joe Clarke · Accepted solution]

It appears your RME database is either corrupt, or its transaction log is corrupt.  See this document on how you can recover both:

https://supportforums.cisco.com/docs/DOC-8796

First try a transaction log recovery.  If that does not work, then you will either need to reinitialize the RME database or restore LMS from a known good backup.

Note: For LMS 2.6, the dbsrv10 will be dbsrv9.

[Brendan O'Flynn]

Hi Joe,

Thanks for the information - the transaction log recovery appears to have sorted the issue.

The only small concern is that while checking everything was OK afterwards I received a FHDbMonitor message on-screen telling me that FHDbEngine process had been restarted on the server, yet the process is running since we completed the transaction log recovery. Screenshots attached. This happens multiple times while accessing CisoWorks.

Many thanks,

Brendan

[Joe Clarke]

I think your DFM FH database is corrupt, and it is periodically crashing.  Follow the instructions in the database doc to renitialize this database by itself.

[Brendan O'Flynn]

Hello Joe,

We decided to validate the database first before re-initialising, as restarting the daemon Manager (later on that day when they first occurred) looked to have gotten rid of the error alerts. The validation of the three DFM databases reported no errors.

We're toying with the idea of leaving as is for the moment, even though we do have backup data to restore afterwards.

Thanks for all your support again,

Brendan