Cisco Support Community
Showing results for 
Search instead for 
Did you mean: 
Community Member

H.323 gateway busy fallback

I have an ICS 7750 w/ 3 mrp 300's all three have 4 fxo ports in them. They all 3 are configured as h323 gateways, and they are all added to a route list in specific order 1 2 3. We have a 9.@ route pattern pointing at that route list. We are having problems on outbound calls when we fill all 4 fxo ports on the first gateway the next person to call outbound gets a busy signal. It is like the router is not sending the call back to the call manager.

We have 4 dial-peer voice X pots w/ preference set 1 2 3 4 then we have a

dial-peer voice X voip w/ preference 5 to send the call back to the call manager.

The call will fail over to the next gateway if the first gateway is offline.

We called TAC and they said that this was not work under H.323 only MGCP

Has anyone run into this sitiuation before.

Thanks in Advance


Re: H.323 gateway busy fallback

Try removing the 5th dial-peer. H.323 gateways cannot do IP-to-IP calls, only IP-to-PSTN calls. The router is failing to send the call to the 5th dial-peer, because this does not work.

Community Member

Re: H.323 gateway busy fallback

We did not have the 5th dial peer until we started having problems then we added the dial peers, but the dial peer did not help w/ all the ports in the router being busy and forwading it to the next router.


Re: H.323 gateway busy fallback

Can you find out what disconnect cause code is being sent by the router? A CCM trace & Q.931 translator should help you find this.

Community Member

Re: H.323 gateway busy fallback

Yea I will get that and get back w/ you on the disconnect cause codes.

Community Member

Re: H.323 gateway busy fallback

You might want to check the service paramter:


Which is defined as:

Specifies when set to true, that if a call being routed through a route list detects the associated "cause" during the release of the call, no re-routing to the next device in the route list is attempted and the call is released with the associated cause.

So if the first GW is returning user busy (from the traces or debug) then this may be your problem.



CreatePlease to create content