Meeting Exchange direct SIP trunk failure.

[BelAirLugosi]

I have a pair of old Meeting Exchange Enterprise 6200 servers which are out of support but I need them to limp on for a few more months.

The linux bridge servers are MX-5.2
The windows servers *were* both Server 2003.
Connected to an Avaya CM6.2 via direct SIP / TCP.

There are a pair of each, a master bridge/crs and a redundant back up bridge/crs.

While I was on leave a month ago someone took the decision to upgrade server 2003 to 2008 on the back up windows box in order to keep the box compliant with house regulations.

They say they tested it for a couple of days, all was good.
The plan was to update the master crs to 2008 also if the first one was successful.

But now the SIP trunk between CM and MX backup is unstable. It fails and shows far-end bypass on the sig grp. If I restart the sig group on the Avaya the tru comes in, I can make calls, then after an hour or so it goes down again. Nothing has been changed on the CM or the Bridge linux box or intervening network.

As I understand it the trunk is between the avaya and the linux bridge so upgrading its associated windows box to 2008 shouldn't affect the trunk.

I have checked the bridge's system.cfg file, all looks good, nothing different to the working bridge.
I have checked its NTP status, that's fine.
I have tried using a different CLAN for sig group near end node name.

Can't work out why the trunk is dropping after an hour or two.

Any ideas?
Should the system work on Server 2008?

Thanks in advance.

 

[Wanebo]

Have you tried Wiresharking the connection?

 

[BelAirLugosi]

No, it might throw something up but I wanted to ask if anyone had had experience on whether updating the windows CRS server could somehow break the trunk to the linux box. I can see how it might mess up something on the CRS but not the trunk itself.

 

[mattKnight]

What happens if you move service to the backup mx server?

You are right though... the Crs server doesn't affect the sip connectivity. It purely passes information to the bridge. I'm surprised That You even got the crs stuff working on server 2012

Take Care

Matt
I have always wished that my computer would be as easy to use as my telephone.
My wish has come true. I no longer know how to use my telephone.

 

[BelAirLugosi]

Matt, the CRS was upgraded from 2003 to 2008, not 2012.
The back up server is the one that has been upgraded to 2008 and now the SIP trunk fails.
The master is still on 2003 and works but is non-compliant.

 

[mattKnight]

We replaced out mx system 2 years ago because of compliance issue withat w2k3 servers.

I'd suspect your mx database is fubarred by the upgraded cars push.

You can use the good credit to push the information to the backup bridge. You need to create a second bridge process on the good credit. There is a bat file that does this.

I'll have a look for the docs if I get a chance

Take Care

Matt
I have always wished that my computer would be as easy to use as my telephone.
My wish has come true. I no longer know how to use my telephone.

 

[BelAirLugosi]

Matt - you have lost me. Grateful if you get a chance to explain that a little?
Why do you think a database would screw the SIP trunk?

 

[mattKnight]

Apologies - I was using my phone to post that and autocorrect corrupted some words... I'll try again!

I'd suspect your MX database is fubarred by the upgraded CRS push.
Unless there is another chnage that you haven't told us about, the most likely scenario is that the CRS OS upgrage is the root cause.

Are you able to create a new conference on the standby CRS and push it to the standby bridge successfully - i.e. make a call using that booking or ad-hoc details? I suspect not, doesn't mean your MX is bust, but the SCAHPI process that syncs between the CRS and the S6200 probably is.

Why that would cause the SIP trunk to be unstable I don't know - but I had odd issues when a log file grew too large (>1Gb). A wireshark would be helpful (I think tcpdump is on the S6200 by default)

You can use the good credit to push the information to the backup bridge. You need to create a second bridge process on the good credit. There is a bat file that does this.
The CRS server can push details to multiple S6200 bridges. There is a piece of interface software that runs on the CRS called SCHAPI which pushes the updated conference details to the bridge. You'd need to start a second instance of this on your primary CRS and configure it to push to you backup bridge. This means that your primary CRS would update both bridges (this may not be what you want - but it may also get you out of a hole)
The way to do this is in the documentation but you will need to runa file called explode.bat in the schapi directory

I'm really sorry - I'm dredging through my memory here - we installed MX 6 years ago and replaced it with AAC8 over 2 years ago.

Take Care

Matt
I have always wished that my computer would be as easy to use as my telephone.
My wish has come true. I no longer know how to use my telephone.

 

[BelAirLugosi]

No need to apologise Matt, grateful for your help. I'll have a play tomorrow now but the good news is the customer might be about to put their hand in their pocket so might just need to limp on for a couple of months.
And you have a better memory than me...