11-09-2010 12:59 AM - edited 03-16-2019 01:49 AM
Hi!
I've got CUCM 5.1 cluster with 2 nodes. Both of them have CFB configured on them and both CFB's are in the same MRG. This MRG is assigned to Device Pool, which is assigned to all phones and gateways. After some period of time I have met a problem with intercluster trunk and I've got to reset CallManager service on the nodes one-by-one to fix this problem. This caused the meetme conference to be disconnencted. Why has the conference disconnected? Is there any way to avoid such behaviour?
TIA!
Solved! Go to Solution.
11-09-2010 06:39 AM
By default your H.323 calls will disconnect unless you have configured on the GW, under your voice-class h323
- call preserve (you can find more information about this command in
the command lookup tool)
Also, please make sure that in CUCM under service parameters (advanced)
you set the below service parameter to true:
- Allow Peer to Preserve H.323 Calls
11-09-2010 03:27 AM
In the device pool associated to your conference bridge, what callmanager group is configured there?
11-09-2010 03:49 AM
This group contains both servers.
11-09-2010 05:00 AM
What type of CFB do you use: the one from IPVMS or hosted on a GW? When you restarted the ccm service, what members were in the meetme conference that were disconnected (pstn call / intercluster trunk / sccp phones).
11-09-2010 06:15 AM
It is software bridges, hosted directly on CUCM servers. During reset in the conference where internal sccp users and 2 or 3 PSTN users, connected through H.323 Gateway
11-09-2010 06:39 AM
By default your H.323 calls will disconnect unless you have configured on the GW, under your voice-class h323
- call preserve (you can find more information about this command in
the command lookup tool)
Also, please make sure that in CUCM under service parameters (advanced)
you set the below service parameter to true:
- Allow Peer to Preserve H.323 Calls
11-09-2010 06:51 AM
Ok, thank you! Will try it. And what to do with MGCP GW?
11-09-2010 11:43 PM
MGCP has call preservation (during failover and failback), more information at the below link:
http://www.cisco.com/en/US/tech/tk1077/technologies_tech_note09186a00806fedbe.shtml
Cheers,
Kristof
11-10-2010 03:08 AM
As far as I remeber, enabling H323 call preserve has one drawback: when phone turns off during conversation (frequent situation in environment with bad cabling and very emotional users) rtp frames are flooding the network unless other side hangs up.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide