InterClusterTrunk CM 3.1 -> CM5 50% of calls drop

[shahedvoicerite]

Hi,

I have an interesting problem.

I have setup a non-gatekeeper InterCluster trunk, between my CM 3.1 and CM 5.0(4).

Approx 50% of calls seem to go through when I dial from CM 5.0(4) to CM 3.

When they dont connect, in the trace, I see

wait_AuConnectErrorInd - send disconn to 1 MXs|<CLID::StandAloneCluster><NID::callmanager5><CT::1,100,15,88.5><IP::10.1.3.2><DEV::Port 3092><LVL::Error><MASK::0800>

In all cases, the called part alerts, and sees the caller id.

Media caps are also exchanged.

Its at the time of media connection, that the call seems to drop.

Having a MTP on one or both ends of the trunk also does not seem to solve the problem.

Attached is a trace file of 2 calls in quick succession.

The first once from 1000 to 489980 succeeds, while the second one, made a few seconds later, fails !!

Any tips would be highly appriciated.

[soupseven]

Hi, there,

Have you found the reason of this problem?

It seems I met the same kind of trouble. I set up an inter-cluster trunk between ccm4.1 and 3.3. And 50% calls got fast busy tone.

could you share some idea with me?

thanks,

[shahedvoicerite]

Hi,

Unfortunately I have not figured this out yet, but some config tewaking seems to reduce the number of fast-busys.

Since my setup is an experimental system only, I am the only one making calls and testing, so I am not sure what the exact scale of the problem is.