h.323 inter-cluster trunk and RTP stream on hairpin call

[44golf58]

The scenario: h.323 trunk between CM clusters in different states. PRI/MGCP Call arrives in cluster A, is transfered to phone in cluster B, and is then transfered back to phone in cluster A. What is the behaviour or the RTP stream between clusters? We are trying to avoid a "hairpinned" (or tromboned) call over the WAN, or at least limit the bandwidth impact.

Thanks,

Dave

[pradeepde]

This URL should help you:

http://www.cisco.com/en/US/products/sw/voicesw/ps556/prod_troubleshooting_guide_chapter09186a008071577e.html

[gogasca]

Cluster 1 <---> Cluster 2

PRIMGCP <-> Phone A Phone B

Call arrives to cluster A

RTP stream now is from GW to Phone A

When call is transferred to Cluster B-

RTP stream now is from GW to Phone B (If you are not using MTP in ICT in Cluster A)

When call is transferred to Phone A.

RTP Stream is from GW to Phone B (If you are not using MTP in ICT in Cluster A)

You can confirm it with debug mgcp packet in the GW and check for MDCX message.

Or with ccm detailed trace in both cluster.

Remember that ICT signalling is H323 like.

HTH

//G

[tomchan]

Hi Gogasca,

Is the last call flow supposed to be --> RTP Stream is from GW to Phone "A in cluster A" (If you are not using MTP in ICT in Cluster A) since the call is transfered back to phone A in cluster A from phone B in cluster B.

Thanks