Strange RTMT message appears few times a day. How to eliminate it?

[christianluven]

HI everybody,

We are running a CUCM 6.1.2 with an IPCC Enterprise and few times a day we get the following message through RTMT:

At Sun Feb 01 14:11:22 CET 2009 on node XXX.XXX.XXX.XXX, the following SyslogSeverityMatchFound events generated: SeverityMatch - Critical CTIManager: 10133: Feb 01 13:11:01.187 UTC : %CCM_CTI-CTI-2-kCtiProviderCloseHeartbeatTimeout: CTI heartbeat timeout; provider closing. App ID:Cisco CTIManager Cluster ID:CM-01-Cluster Node ID:cm-01 Process ID:JTAPI[28]@UNIX-PC5 Process ID:24784 Process Name:CtiHandler

We were faced some major Problems with our Callcenter in the last week where we were forced to restart the ICM-Services and the CTIManager-Service on the CUCM.

Is it possible that the problem is linked with this error message?

How to eliminate this error message?

greetings

Christian

[Chris Deren]

Absolutely, as you can see this is CTI error which affects applications such as UCCE. What do you see in the event viwer on UCCE PG at that time? This error typically indicates network issues as communication between CM and PG may be interrupted, also make sure your JTAPI versions are the same on CM and PG, as well as that the NIC bindings are properly set on the PG server.

HTH,

Chris

[christianluven]

HI Chris,

Thanks for your reply!!

In the Router event log i see nothing.

The components of the UCCE are divided over two sites PG's at site A and Rogger at site B. Sometimes the WAN-Connection between the two sites seems to be overloaded and the Voice Environment doesn't really use the backup-connection.

Than i have to restart the CTI-Manager service at the CM at the Site where is no PG located, and also the ICM-Services.

Strangely and fortunately the error message is not simultanous with the non working UCCE.

I get the error message few times a day without impact on the daily business.

I would like to know which connection is broken for a short time i think there are multiple servers involved.

thanks an greetings

Christian

[Chris Deren]

Christian,

It appears your environemnt may not be properly configured for redundancy. First of all I would validate that the PGs point only to the local JTAPI Providers (CTI Managers on CM servers) and not the ones across WAN.

You should never have to reset anything manually, after a network issue the system should come into stable state on its own.

HTH,

Chris

[christianluven]

the bad thing is that we have only PG's at one site, working for both sides....

So this is the only way it works.

greetings Christian

[Chris Deren]

Christian,

That is not supported deployment as your PG CM PIM has to be collocated with CTI Server, depending on the version it is the same case with IPIVR (Cisco keeps changing SRND to say VRU can be across WAN or not).

My recommandation is first to resolve this and have a supported deployment model.

HTH,

Chris