Can a failover incomming Dial-plan be configured?

jaydien1358 · ‎08-26-2010

Hello,

Had a disappointing situation occur to our in-house UC520 this week. Sometime on Saturday 8/21, our office experienced a power failure. This failure caused an internal module that CUE runs it's image file off of to become corrupt. Hence voicemail and AA no longer function.

The crappy part is that outbound calls were able to be made as normal.

Long story short, we did not know our AA wasn't picking up inbound calls until 4 days into this. And I imagine some of our customers were unable to reach us, unbeknownst to us.

My question is, Is there any way to have a failover inbound dial-plan for situation where the AA is not reachable. Maybe have all call forward to a blast group that way we can avaid this in the future?

Thanks.

Brian

www.jaydien.com

antyeung · ‎08-26-2010

This is not possible from CCA. However, it's possible from the CLI (and not pretty) but you would have to test it out.

Assuming this is your dial-peer to CUE-

dial-peer voice 2000 voip

destination-pattter 399

session protocol sipv2

session target ipv4:10.1.10.1

dtmf-relay sip-notifcy

codec g711u

no vad

Add another dial-peer that routes the call to itself w/ the same destination-pattern

dial-peer voice voip

destination-pattern 399

incoming called

ipv4:x.x.x.x

codec g711u

no vad

pref 1

translation-profile outgoing xlateToHuntPilot

In the translation profile translate the called number 399 to the blast group pilot.

What I think should happen is when you call 399 and it gets sent to CUE and fails the call would attempt to route to the second dial peer where you translate the called number to itself. Since the called number was xlated to the blast group pilot on the incoming it should route to the blast group.

Thanks,

-Anthony

jaydien1358 · ‎08-27-2010

Thanks but that isn't as clean as I was hoping.

Is this something that has been tested at all? Does anyone else have an idea of how to do this cleannly?

Thanks.

-Brian

www.jaydien.com

antyeung · ‎08-29-2010

This has not been tested, but it should work. You'll need to test it to verify that it does what you want.

The only way for CME to know that CUE is in an inoperable state is for CME to extend a call to CUE and for it to fail. You can make use of the "monitor probe icmp-ping" feature configurable under the dial peer where CME would send ICMP pings to the destination you specify. If ICMP ping fails the dial peer status would change. However this may not be what you want since CUE may still be pingable (interface UP), yet the CUE software may be inoperable. Also, you'll still need to configure a secondary dial peer where you'll need to loop it back to itself.

AFAIK, there isn't a cleaner way to achieve failover w/out manual intervention.

If others have ideas/suggestions I would be interested in hearing them.

Thanks,

Anthony