CUCM 10.5: Cisco Prime LM DB service down

[panjwani80]

I am seeing lot of RTMT alerts for Cisco Prime LM DB service down in CUCM 10.5. I don't see any service related to it on serviceability page.

service won't start via CLI either. How do I start his service so that RTMT alerts stops

Cisco Prime LM DB[STOPPED]  Component is not running

admin:utils service start Cisco Prime LM DB
Service Manager is running
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Component is not running
Cisco Prime LM DB[NOTRUNNING]

[m.batts]

Sounds like you're suffering from the following bug. You need 10.5.SU1

ServM sees PrimeLM DB Service as not started after upgrade to 10.5(1)SU1

CSCur32516

Description

ServiceManager reports a Service Down Alarm for "Cisco Prime LM DB" and the "utils service list" output shows that that the service has not started. Prime LM appears to be functional when logging into the Prime LM Admin page.

The Cisco Prime LM DB is seen as not running from the service list output:

Cisco Prime LM DB[STOPPED] Component is not running

The following error is continuously logged in the Application Log:

Oct 21 16:42:58 CUCM105Sub1 local7 2 : 173: CUCM105Sub1: Oct 21 2014 20:42:58.796 UTC : %UC_SERVICEMANAGER-2-ServiceFailed: %[ServiceName=Cisco Prime LM DB][Reason=Service stopped abruptly][AppID=Cisco Service Manager][ClusterID=][NodeID=CUCM105Sub1]: Service terminated.

Symptom:
ServiceManager reports a Service Down Alarm for "Cisco Prime LM DB" and the "utils service list" output shows that that the service has not started. Prime LM appears to be functional when logging into the Prime LM Admin page.

The Cisco Prime LM DB is seen as not running from the service list output:

Cisco Prime LM DB[STOPPED] Component is not running

The following error is continuously logged in the Application Log:

Oct 21 16:42:58 CUCM105Sub1 local7 2 : 173: CUCM105Sub1: Oct 21 2014 20:42:58.796 UTC : %UC_SERVICEMANAGER-2-ServiceFailed: %[ServiceName=Cisco Prime LM DB][Reason=Service stopped abruptly][AppID=Cisco Service Manager][ClusterID=][NodeID=CUCM105Sub1]: Service terminated.

Conditions:
Caused by CSCuq15891

Workaround:
If the node is not being used to manage licenses, Prime LM can be removed from that node by running "license management system remove", apply CUCM 10.5(1)SU1a [10.5(1.11901-1)] which is posted to CCO and contains a fix for this issue or otherwise contact TAC for a workaround.

Note: The command "license management system remove" is valid only for 10.5(1)SU1 and later and forces a server reboot.

Further Problem Description:

[m.batts]

Sorry 10.5 su1a

[Janderso6]

Is there a resolution for this problem.

We have a publisher and 4 subscribers. All I was doing was a routine reboot of our servers starting with the publisher and followed by each subscriber.

Call Manager version: System version: 10.5.1.10000-7

Took screen shots of all services using utils service list and the command Cisco Prime LM DB[STARTED] was activated on all them before rebooting each server one at a time.

After all services started on publisher checked:

Publisher - Cisco Prime LM DB[STARTED]
Subscriber - Cisco Prime LM DB[STOPPED] Component is not running
Subscriber - Cisco Prime LM DB[STOPPED] Component is not running
Subscriber - Cisco Prime LM DB[STOPPED] Component is not running
Subscriber- Cisco Prime LM DB[STARTED]

Now my email wont stop spamming from 3 out of the 4 subscribers with [RTMT-ALERT-StandAloneCluster] CriticalServiceDown

I tried to run the following command:


admin:utils service start Cisco Prime LM DB
Service Manager is running
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Cisco Prime LM DB[STARTING]
Component is not running
Cisco Prime LM DB[NOTRUNNING]

Why does it work for the publisher and one subscriber? I can login into Cisco Prime License Manger Version: 10.5.1.10000-9 and everything appears fine.
 

Please help!!!

[m.batts]

You need to upgrade to 10.5.1 SU1a

[dallem_calltower]

I ran into the same issue and simply did a reboot of the same server and all services came back without issue.

[Dennis Mink]

Just to comment on this, I have seen this issue on 10.0. as well.

[Hard_Boiled_Greg]

I also started to see this issue. However, it is on a CUCM running version 10.5.2.11900-3. I have attempted to restart the service from the CLI:

admin:utils service start Cisco Prime LM DB
  Service Manager is running
  Cisco Prime LM DB[STARTING]
  Cisco Prime LM DB[STARTING]
  Cisco Prime LM DB[STARTING]
  Cisco Prime LM DB[STARTING]
  Cisco Prime LM DB[STARTING]
  Cisco Prime LM DB[STARTING]
  Cisco Prime LM DB[STARTING]
  Cisco Prime LM DB[STARTING]
  Cisco Prime LM DB[STARTING]
  Cisco Prime LM DB[STARTING]
  Cisco Prime LM DB[STARTING]
  ...
  Component is not running
  Cisco Prime LM DB[NOTRUNNING]

I have also rebooted the server. This happens to be a single node cluster. The error was still present.

I eventually opened a service request with TAC. Cisco actually had me completely disable the Cisco Prime LM DB service on the cluster by issuing the command "license management system remove" from the CLI. Since we use a centralized Prime License Manager to manage our licenses we don't require the localized Prime LM DB service on the cluster. The odd thing is that we had been running 10.5(2) on this cluster for at least five months, and the issue just surfaced recently. Cisco was not able to tell me what triggers the Prime LM DB issue, and the TAC engineer seemed a bit surprised to see it surface in version 10.5(2). Regardless, the issue in my case is now resolved simply by completely disabling the service.

[Venkatesh Perumal]

https://supportforums.cisco.com/discussion/12740296/cisco-elm-db-service-not-running#comment-11047576

[Hard_Boiled_Greg]

The support discussion referenced deals more with disabling RTMT alerts as opposed to correcting the Prime LM DB issue we are seeing. I don't need to disable the alert, and since the actual alert is clumped in the "CriticalServiceDown" category I certainly don't want to disable that. Cisco actually had me completely disable the Cisco Prime LM DB service on the cluster by issuing the command "license management system remove" from the CLI. Since we use a centralized Prime License Manager to manage our licenses we don't require the localized Prime LM DB service on the cluster.

[Mohammed-A.R.A]

Is this resolved? Ran into same issue.