SBC / SM

[wpetilli]

We had an incident with an internet provider serving our datacenter A, where I have a SBC / SM for remote workers. Our DNS probe repointed the traffic to our datacenter B SBC. The endpoints successfully registered, however, I noticed 2 major things:
1. I was able to receive calls on the remote endpoints. That was a good thing. These come in via PSTN to CM and have OPS entries for the SIP endpoints that traverse the SIP trunks to SM. However, I was unable to make calls. I received a fast busy on both internal and external calls.

2. While I had successful registration, I had no PPM data and had no bridged appearances.

I know these issues are a combination of things. What's the best way to unpeel this onion?

Once the datacenter a circuit issue got resolved and DNS failed back, things worked again.

 

[wpetilli]

I didn't get enough time to really dig into this, but I can probably assume that the endpoint registered to the SM in datacenter A, as it's the primary SM for that endpoint and was available. Not sure if that matters or not.

 

[kyle555]

Is the authoritative domain of the network region for the trusted SBC interface at DC2 set to the same thing as the authoritative domain for the NR for trusted SBC interface at DC1?

 

[wpetilli]

My SIP trunks from CM to each of these SM's both use the same NR and both have the same domain set.

 

[G van Hamburg]

Normally we use SBC-A + SM1 in DC1 and SBC-B + SM2 in DC2. The remote workers register to both SM’s and when the users primary SM fails, the secundary SM handle’s PPm and calls. I can’t see you setup from here but could it be something is wrong there?

Freelance Certified Avaya Aura Engineer

 

[kyle555]

Yeah but are both A1 IPs in the network map as belonging to the network region you expect the phones to register to?

 

[fondog2]

Just curious but does this have a proxy route set for those locations and are all the SIP trunks added to it?

 

[jimbojimbo]

This is not the way I would have configured the system however, you should still be able to troubleshoot. Use an Equinox/Workplace client and set the server address to the SBC IP in the secondary datacenter. You should be able to run a traceSBC and include PPM in the data to see if you are getting anything.

The SBCs are independent so check the PPM proxy settings of the SBC in the secondary datacenter.

You may also want to go with two DNS entries for the SIP registration. One for the primary and one for the secondary data centers. For AADS you will still want to use an external load balancer (assuming you have an AADS server/cluster in each datacenter).