H.323 secondary gateway issue when E1 service fails

Unanswered Question
Aug 21st, 2007

Hi,

We've had a problem today with one of our remote sites. The E1 service failed and the users were unable to receive inbound external calls (as expected) or to make outbound external call (not expected).

The Call Manager 4.0 cluster has two H.323 gateways for external calls for each remote site (the primary is routed via the remote site E1 service and the secondary via the CCM site).

The site has been configured so that if all ISDN ports are busy new external calls overflow to the secondary gateway. This has been tested and works fine. SRST has also been configured on the router and tests OK.

The problem that happened today appears to be that when the E1 service failed the calls were still being routed to the primary H.323 gateway and not being sent to the secondary gateway. This was becaus the loopback address that binds the gateway address was still up up.

Is there any way to configure the router or the call manager so that the above situation does not happen and that if the E1 service fails new calls are routed to the secondary gateway.

Some configs below.

Loopback config for H.323 gateway

interface Loopback0

description Loopback Address

ip address 10.x.x.x 255.255.255.255

h323-gateway voip bind srcaddr 10.x.x.x

sh ip int br

Serial0/0/0:15 unassigned YES NVRAM down down (E1 service down)

Loopback0 10.x.x.x YES NVRAM up up (Loopback interface up)

Route/Hunt list Configuration (in order of priority)

Remote_RG[non-QSIG] (Remote site)

Local_RG[non-QSIG] (CCM site)

Note: When the route list priorities were changed manually the remote site could make external outbound calls via the secondary gateway.

Thanks in advance for help provided.

Regards,

Bryce.

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Loading.
Anonymous (not verified) Mon, 08/27/2007 - 12:51

In some situations, publisher and subscriber are used as the secondary and primary servers respectively. In this case the lower preference should be given to the subscriber because it is the Cisco CallManager server designated for call handling, while the publisher is designated to handle both the SQL database and the LDAP directory. If there is any problem with the subscriber, the call should be routed to the publisher, which can also route calls.

fbender Mon, 08/27/2007 - 20:56

Hi,

one option might be to bind the ip address to the E1(serial) instead of the loopback. So if the E1 goes down the IP will no longer be reachable.

Regards

Frank

boxhallbr Wed, 08/29/2007 - 21:16

We think we've found the answer to the issue however we haven't been able to test it yet.

On the Call Manager in the Service Paramater Configuration page you chose true for Automated Alternate Routing Enable*

Will advise how it went. Thanks for help.

Actions

This Discussion