I have a new 2851 running 12.3911)T10
1 MFT T1 in the MBRD slot 3
CM 4.1.2 with a publisher and sub
I will past the config in here, but here is the issue, the Sub as the Primary will not register even though the Pub is fine. (see "sh ccm" output) I have verified:
--DNS name and gateway name in CM and router both include "spps.com"
--stipped and re-applied "mgcp" global command
--ran DB Helper to make sure database was replicating
I am all out of ideas on this.
MGCP Domain Name: VR01-1930.spps.org
Priority Status Host
Primary Down 10.1.8.15
First Backup Registered 10.1.8.16
Second Backup None
Current active Call Manager: 10.1.8.16
Backhaul/Redundant link port: 2428
Failover Interval: 30 seconds
Keepalive Interval: 15 seconds
Last keepalive sent: 14:27:55 DST Jun 23 2006 (elapsed time: 00:00:10)
Last MGCP traffic time: 14:27:55 DST Jun 23 2006 (elapsed time: 00:00:10)
Last failover time: 13:40:55 DST Jun 23 2006 from (10.1.8.15)
Last switchback time: 12:19:15 DST Jun 23 2006 from (10.1.8.16)
Switchback mode: Graceful
MGCP Fallback mode: Not Selected
Last MGCP Fallback start time: None
Last MGCP Fallback end time: None
MGCP Download Tones: Disabled
--More-- Backhaul Link info:
Link Protocol: TCP
Remote Port Number: 2428
Remote IP Address: 10.1.8.16
Current Link State: OPEN
Packets recvd: 0
Recv failures: 0
Packets xmitted: 0
Xmit failures: 0
PRI Ports being backhauled:
Slot 0, port 0
Configuration Error History:
FAX mode: disable
SH run vitals:
Current configuration : 2487 bytes
service timestamps debug datetime msec
service timestamps log datetime msec
no service password-encryption
ip domain name spps.org
no ftp-server write-enable
isdn switch-type primary-ni
ccm-manager redundant-host 10.1.8.16
no ccm-manager fax protocol cisco
ccm-manager config server 10.1.8.15
controller T1 0/3/0
pri-group timeslots 1-24 service mgcp
description 1930 Como PRI
description Connection to VOICE NET on 1930-Como Switch
ip address 10.1.8.10 255.255.254.0
no ip address
isdn switch-type primary-ni
isdn incoming-voice voice
isdn bind-l3 ccm-manager
no cdp enable
ip route 0.0.0.0 0.0.0.0 10.1.8.1
mgcp call-agent 10.1.8.15 2427 service-type mgcp version 0.1
mgcp dtmf-relay voip codec all mode out-of-band
mgcp rtp unreachable timeout 1000 action notify
mgcp package-capability rtp-package
no mgcp package-capability res-package
mgcp package-capability sst-package
no mgcp package-capability fxr-package
mgcp package-capability pre-package
no mgcp timer receive-rtcp
mgcp sdp simple
mgcp fax t38 inhibit
mgcp rtp payload-type g726r16 static
mgcp bind control source-interface GigabitEthernet0/0
mgcp bind media source-interface GigabitEthernet0/0
mgcp profile default
As mentioned that network connectivity is fine.What abt the phones do the they get registered to PUB.Can you make calls between phones which are connected to seperate CCM.
Can you verfiy the version of softwares running on both the servers.
Try restarting the Callmanager service in Publisher. Also make sure the Database layer service is running.
Any errors related to MGCP or devices on the error log.
In the config you posted, the dial-peer command is missing. Do you have that in the router ?
dial-peer voice 1 pots
You are missing 2nd backup host name
try with below configuration :
ccm-manager redundant-host 10.1.8.15 10.1.8.16
Let me know if this helps.
Thats not required. His pub is .16 and sub is .15. His mgcp call-agent has .15 specified. He has a ccm-manager redundant-host with .16 specified too.
Is the dial-peer needed for registration? I had not finished the config b/c the gateway had notr registered.
Also, the phones are all working fine off both CM's as the enviornment is pre-existing.
I did see in DBHelper that the Pub DNS is listed as CCM1.spps.org while the Sub is listed simply as CCM2
When I look at the box itself they both have the spps.org defined in the fully qualified domain extension (both are in a simple workgroup named WORKGROUP) so I'm not sure why DBHelper is reporting it differently. I do get green smilies though, not Mr. YUK's SO I don't know if this is really a problem, but considering how MGCP reacts with not ergistering on domain name mismatches AND the Sub is the one listed with out the spps.org, it might have something to do with it.
I am simply adding a new gateway so I can get rid of the existing H323 T1 gateway. So all phones and calls work fine right now. This gateway is not in the equation yet.
I asked the client to restart his Sub at the end of the day friday to see if that would fix the issue. I thought it might be a service on one of the servers, but wasn't sure which service would affect MGCP registration. The Pub must be ok since it registers but the Sub does not (see "sh ccm" output)
great feedback so far guys, I hope this extra info sparks some ideas. I'll be opening a TAC case on Monday, but I'd like to get it resolved before having to do that. It's such a basic set-up that it is frustrating the heck out of me to see the Sub listed as "Down" in the Gateway.
One thing I have noticed, is that when configuring mgcp manually (no ccm-manager config) I have needed to add the dial-peer to get it to register initially, but removing it after that caused no issues.
Its very strange. Anyway, I was just curious to see if anyone else has experienced this.
Again, I stress that IT IS NOT NEEDED as shanky said, but initially adding it has caused my pri's to register to callmanager immediately (and then can be removed after that as MGCP controls the d channel and this is why its not needed for call routing).
adignan - berbee
I am not sure if you have checked this, but If you have a ip domain name specified, when you add your gateway to CCM you have to specify the domain name.
Hope this helps
thank you eloy, I have checked this and tried it both ways in CM and on the gateway.
Added info - I also stripped the gateway config clean and set it all back in to see if having it defined in CM first made any difference. It did not.
On the GW, can you enable:
deb mgcp packet
deb ccm events
no mgcp - wait a couple of secs for the debugs to stop, then
then give a few seconds fo the registration to complete, then do "unde all", and sh ccm ( jsut to confirm it's registered fine)
I'll take a look at the traces to see what they show.
Wild goose chase......The client had applied MS patches to the subscriber which was causing the services to flake out. We uninstalled the patches and everything came up fine.
Thanks for all your help.