Author Topic: ZBTSI Protocol  (Read 1484 times)

Offline ralph

  • Mitel Forums Admin
  • Hero Member
  • *****
  • Posts: 5767
  • Country: us
  • Karma: +469/-0
  • Published Author: http://amzn.to/2dcYSY5
    • View Profile
ZBTSI Protocol
« on: February 05, 2016, 03:39:44 PM »
This apparently is giving me grief on some of my 3300s.
I know most of you will never have heard of this protocol (I didn't until today) but I'm throwing this out there in case someone has.

Has anyone see an explanation, and I mean bit by bit, of the ZBTSI protocol?
It does the same thing as B8ZS but differently.
You *might* see this as an option on a T1 testing tool.    Mine doesn't have it.  Newer ones might.

Here's what happens:
I have a T1 with 24 PLARs on them.  (E&M Immediate Start)
If ZBTSI is enabled anywhere along the line from here to the CO or far end, it makes my 3300 appear to make all 24 channels go off hook outbound at the same time even though it isn't.
It's taken a long time to come to this conclusion.  I actually had to talk to a guy who wrote the book on T1s (Literally) and he clued me on the protocol.   I just haven't see a diagram or bit by bit explanation for what it's doing.

Ralph



Offline johnp

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 2201
  • Country: us
  • Karma: +66/-0
    • View Profile
Re: ZBTSI Protocol
« Reply #1 on: February 06, 2016, 07:07:14 PM »
I never have heard of that. Did a google and found this http://lyle.smu.edu/~levine/ee6302/chanb4.pdf look at pages 10 through 12

Offline ralph

  • Mitel Forums Admin
  • Hero Member
  • *****
  • Posts: 5767
  • Country: us
  • Karma: +469/-0
  • Published Author: http://amzn.to/2dcYSY5
    • View Profile
Re: ZBTSI Protocol
« Reply #2 on: February 07, 2016, 06:50:57 PM »
That is awesome!
So much further than I was able to go.
If there is a mismatch in protocol somewhere B8ZS vs ZBTSI then what happens is ZBTSI shifts the bits one over.
On PLARs the signaling bits are all 1s at idle.   Shift a zero over and bang!  outbound seizure! on 24 channels.

Here's a test I did:
I can duplicate the issue on a particular T1 by making several outbound calls and disconnecting before far end answers.
I've put a Adtran Total Access unit between the carrier and the PBX that will generate syslogs to show me what channels are seizing in what direction.
When an 'event' happens the syslogs show all 24 channels seizing outbound from the Mitel.  The Mitel never shows outbound calls.  It may show incoming calls on some channels.
To complicate things if you move the T1 to another port, the problem follows the T1. 
This of course points to the T1 but every tool I have points to the Mitel so now I have to figure out what's really going on because everybody says "It's not us."

I next put an Adtran TSU120 between the Adtran TA unit and the carrier and used it to isolate channel 24 from the Mitel just to see if anything on the T1 channel itself was causing it.   When I tested it I still saw the Mitel seizing channel 24.   ***remember that channel 24 was now isolated from the Mitel ***

I ended up speaking by phone to William Flanagan who wrote this book: [urlwww.amazon.com/Guide-T-1-Networking-Flanagan-Paperback/dp/B011DC4KJK/ref=sr_1_4?ie=UTF8&qid=1454888542&sr=8-4&keywords=william+Flanagan+t1]Guide to T-1 Networking: How to Buy, Install & Use T-1 From Desktop to Ds-3[/url]

He's the one that told me about the ZBTSI protocol and the bit shift but I haven't been able to find anything on it all until you found it for me.  (Thanks, by the way)

Unfortunately it's not helping me much since the carrier has pointed out that it's a Verilink proprietary protocol and they can't even turn it on in their routers.

One thing I did find is that it is a "European Proposal".   Since at least 2 of my T1s with the issue have DS0s extended to London this sorta makes me think there might be a little "smoke" here.
I just haven't been able to prove it yet.

So I'm hoping some of you over in Europe may be familiar with this?  Seen it?

Ralph







 






 

Sitemap 1 2 3 4 5 6 7 8 9 10