[asterisk-users] Calls on E1 TDMoE span are dropped at random

[Steve Totaro]

On Tue, May 13, 2008 at 9:57 AM, Florian Hackenberger
<f.hackenberger at chello.at> wrote:
> On Tuesday 13 May 2008, Steve Totaro wrote:
>  > You can be shot several times and not die.  I would try
>  > resetinterval=never just to be able to to say "Not the problem"
>  > rather than "Probably not the problem.
>  I'll do that, although I'm pretty sure that the setting is not the
>  problem as the yellow alarm occured quite often a few minutes after
>  restarting asterisk and the default is 3600 seconds.
>
>
>  > PRI debug info would be a great help too.
>  The log I sent in the original message contains pri debug messages. I
>  just had another look at it.
>
>  Thanks for your help!
>
>
>
>  Cheers,
>         Florian
>
>  --
>  DI Florian Hackenberger
>  florian at hackenberger.at
>  www.hackenberger.at
>

Ah, didn't even know you could add attachments to postings to the list.

It is a little hard to read and quite a bit of info that may or may
not be a problem.  Does this happen enough, or do you have enough time
to sit there and catch the exact log output when it happens?

If you can comment out the spans you are not using, that would reduce
a bit of output (I assume you have a single E1 and some POTS (although
I don't see them configured,  but you said in your initial posting
that you could dial out on POTS).

"ERROR[7968]: chan_zap.c:8176 zt_pri_error: !! Got reject for frame
120, retransmitting frame 120 now, updating n_r!" and "ERROR[7968]:
chan_zap.c:8176 zt_pri_error: !! Got I-frame while link state 2 -- Got
UA from network peer  Link up." looks suspicious.  Maybe Red-fone
could give you some insight on these errors.

If you can narrow it down then I am sure someone can better.  Again,
comment out spans not in use, set your verbose to 0, and turn on PRI
debugging and try to catch only the event/s that correlate with the
calls being dropped.

I saw Red-fone's products at Astricon, they looked great for failover.

Can you describe exactly how you are utilizing it, including LAN/WAN,
switches, ping times, and other network central details.  TDMoE adds
the E (ethernet) component to troubleshooting and I think do to this,
it may be very fragile depending on network conditions.

Don't make the mistake of just focusing on Asterisk and Zaptel in your
troubleshooting process.

Thanks,
Steve Totaro