05-22-2011 09:46 PM - edited 03-16-2019 05:05 AM
I have CUCM 7.1.4 and remote IOS MGCP Gateway, connected via E1 to PBX.
Everything works fine untill WAN link temporarily fails. After link recovers, MGCP gateway stops working.
# sho mgcp
MGCP Admin State ACTIVE, Oper State ACTIVE - Cause Code NONE
MGCP call-agent: 10.96.36.2 2427 Initial protocol service is MGCP 0.1
CUCM show what gateway is registered okay, but calls get rejected.
Call tracing revealed this message:
05/23/2011 09:58:56.220 CCM|ConnectionManager - wait_AuDisconnectRequest ERROR:NO ENTRY FOUND IN TABLE,CI(18054240,0),dcType=1,IFCreated(0,0),PID(0-0,0-0),IFHandling(0,0),MCNode(0,0)|<CLID::StandAloneCluster><NID::10.96.36.2><CT::1,100,39,1.49539036><IP::10.96.41.18><DEV::SEP1C17D3404C98><LVL::Error><MASK::0800>
issuing
(config)# no mgcp
(config)# mgcp
fixes problem untill next WAN failure
trace log attached
Call made at 09:58:55.959
calling number 8489
called number 0040
05-23-2011 12:03 AM
Switch to H.323 or SIP, you will enjoy more features and less trouble,
05-23-2011 01:18 AM
Yeah.
Then switch to Asterisk.
05-23-2011 01:27 AM
Very wrong analogy, get informed before talking.
05-23-2011 04:20 AM
Hi,
What version of IOS you running?
There have been bugs that affect fail back to MGCP after a switchover.
Have you done a search on the cisco.com bug toolkit?
Cheers,
Tim
05-23-2011 04:31 AM
0tim.smith wrote:
Hi,
What version of IOS you running?
There have been bugs that affect fail back to MGCP after a switchover.
Have you done a search on the cisco.com bug toolkit?
Cheers,
Tim
Tried different versions. Right now it's Cisco IOS Software, 2800 Software (C2800NM-ADVENTERPRISEK9-M), Version 12.4(24)T5, RELEASE SOFTWARE (fc3). Router 2811.
05-23-2011 05:10 AM
Ok so quite recent.
What about your IOS configuration?
If there's nothing obvious, then your best bet would be a TAC case.
I also agree that H.323 and SIP are usually nicer, but obviously that's a design change. And MGCP should work - it does normally switch back automatically!
However - 1 point to consider is call preservation. You'll drop all your calls when switching call control from MGCP to local in SRST mode. Not a good look normally.
Cheers,
Tim
05-23-2011 05:23 AM
Call preservation is really the lesser of H.323 advantages - as it requires the non TAC supported disabling of H.225 keepalives.
The real advantages are actually others, robustness, flexibility, serviceability, config cleanliness, call blocking, TCL/ICR script support, and some more.
05-23-2011 05:47 AM
Off topic - but have you tried this out?
Cheers,
Tim.
05-23-2011 10:28 PM
What about your IOS configuration?
All by the book.
(config)# ccm-manager config server x.x.x.x
(config)# ccm-manager config
If there's nothing obvious, then your best bet would be a TAC case.
I did already. They took time to investigate.
I also agree that H.323 and SIP are usually nicer, but obviously that's a design change.
I'd prefer to try and fix problem first. Having centralized management is one of reasons we've spent a fortune bying CUCM.
10-12-2012 01:06 PM
Ultair,
Did you ever get this resolved? We are having the same issue with our MGCP not recovering too.
CCM - System version: 8.6.2.20000-2
2811 Router - Version 12.4(24)T6
Thanks!!
10-23-2012 08:16 PM
No, switched to SIP. Easy to debug, easy to trace.
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