Has anyone once achieved thsi?
We are trying to connect the CME SIP trunk ( 12.4.22T advipservices ) on a 2811 to a VX900
We can call from SIP phone registered on VX to phone registered on cisco CME but not back so it looks like a dial-peer issue
though I can't think of anything . There is no dial tone on SIP phone at CME ( SPA922 )
 
Any suggestions?
 
Here is our config:
 
voice service voip
 allow-connections sip to sip
 sip
  registrar server expires max 600 min 60
!
!
voice class codec 1
 codec preference 1 g729br8
 codec preference 2 g729r8
 codec preference 3 g711alaw
!
voice register global
 mode cme
 source-address xx.xx.xx.xx port 5060
 max-dn 8
 max-pool 8
 authenticate register
 dialplan-pattern 2 690.... extension-length 4
 tftp-path flash:
 create profile sync 0007272453516527
!
voice register dn  1
 number 6902222
 allow watch 
! 
voice register pool  1
 id mac [mac-address of client]
 type 7970
 dtmf-relay sip-notify
 voice-class codec 1
 username xxxx password xxxxx
!
 dial-peer voice 1 voip
 description ** SIP TRUNK **
 destination-pattern 686....
 voice-class codec 1
 voice-class sip dtmf-relay force rtp-nte
 session protocol sipv2
 session target ipv4:[IP of remote VX SIP server]
 dtmf-relay rtp-nte
 no vad
  !
telephony-service
 em logout 0:0 0:0 0:0
 max-ephones 10
 max-dn 20
 ip source-address xx.xx.xx.xx port 2000
 load 7960-7940 P00308000500
 dialplan-pattern 1 690.... extension-length 4
 max-conferences 8 gain -6
 transfer-system full-consult
 create cnf-files version-stamp Jan 01 2002 00:00:00
!
 
 
 

I'm sure the original poster has long figured this out, but recently I've run into a similar issue and found the cause to be the pattern on the VX900 not being syntactically correct for the SIP messaging.
 
For example, I had {$$$}{$$$$} for incoming matching, but in order to preserve the id of "e.164num@ipaddress" the pattern should be {$$$}{$$$$}@+

 
I had a problem like this also. My problem was the VX900. The CME IP source address was a loopback (192.168.20.1) and the CME router was the default gateway (192.168.1.1) of the network the VX was on. I could call from a PBX switch through the VX to the CME, but I couldn't call back. The problem was the VX was seeing the default gateway as the inbound calling party not the loopback address. To fix this I made the SIP inbound the default gaetway and everything worked fine. I guess you could also just make the CME IP source address the default gateway. I hope this helps someone, because this is the only place I could find anything about making a VX and a CME communicate. If you go to the CLI of the VX Enable->Trace all (Trace none, to turn off) and then call you can see what the issue is.