[Asterisk-Users] RE: Dropped calls continued

[whois wes]

Dan,

We are also using E&M Wink T's, and have been fighting call drops for
several months....our config is as follows:

8 total servers, between 2 and 4 T's per server
Dell PE 2850s, 4GB RAM, RAID 1 HD's
Asterisk 1.2.4, Zaptel 1.2.5

We initially had Digium TE410P's in all of the boxes, and between
constant drops and random hard lockups of the servers every few days,
we decided to switch to Sangoma.

Since then, not a single lockup, and the call drops went from 1-2% of
the calls to under 0.05%.  Out of 7000 calls the first day on the
Sangoma, we had 3 drops.  Day before, over 100.

This was after isolating the digium card on it's own IRQ, disabling
every non-essential onboard device, turning off ACPI, and several
other attempts to stabilize the systems - nothing worked, and nobody
could come up with a reason why.

Moving to Sangoma fixed just about every issue we've had.

Not sure if this helps you at all, but if you were at all able to get
your hands on a Sangoma based card, you might give it a shot.

Wes

On 6/15/06, Dan Elder <isbeen at izmedia.com> wrote:
> sorry, forgot to include some relevant info in my rush to find an answer:
>
> This is * 1.2.8
>
> Non PRI T1 (e&m)
>
> New T1 recently installed, going into a Digium 4pt T1 card..
>
> Zaptel.conf:
> span=1,1,0,esf,b8zs
> e&m=1-24
>
> Zapata.conf
>
> usecallerid=yes
> usecallerid=yes
> hidecallerid=no
> context=from-pstn
> callwaiting=yes
> callwaitingcallerid=yes
> threewaycalling=yes
> transfer=yes ; allows flash transfers
> cancallforward=yes
> language=en
> ;signalling=fxs_ks
> signalling=em_w
> rxwink=300              ; Atlas seems to use long (250ms) winks
> group=1
> immediate=no
> faxdetect=no
> callgroup=1
> channel=>1-24
>
> -----Original Message-----
> From: Dan Elder [mailto:isbeen at izmedia.com]
> Sent: Thursday, June 15, 2006 1:47 PM
> To: 'asterisk-users at lists.digium.com'
> Subject: Dropped calls continued
>
> Hi All... Well, I'm still experiencing LOTS of dropped calls since
> installing the new (non pri) T1 here... I keep noticing a few things in the
> logs when this happens, namely the "Wink/Flash" statements and the "Didn't
> get a frame" messages...
>
> Anyone got any ideas on if this is a telco issue, a wiring issue, or an
> asterisk issue? Been trying to track this down via all 3 paths, but not
> having any luck.
>
> chan_zap.c: Exception on 18, channel 2
> DEBUG[11577] chan_zap.c: Got event Wink/Flash(3) on channel 2 (index 0) Jun
> 15 13:33:00
> DEBUG[11577] chan_zap.c: Ignoring wink on channel 2 Jun 15 13:33:01
> DEBUG[11577] chan_zap.c: Exception on 18, channel 2 Jun 15 13:33:01
> DEBUG[11577] chan_zap.c: Got event Hook Transition Complete(12) on channel 2
> (index 0) Jun 15 13:33:03
> DEBUG[11577] chan_zap.c: Exception on 18, channel 2 Jun 15 13:33:03
> DEBUG[11577] chan_zap.c: Got event Dial Complete(9) on channel 2 (index 0)
> Jun 15 13:33:03
> DEBUG[11577] chan_zap.c: No echo cancellation requested Jun 15 13:33:12
> DEBUG[11577] chan_zap.c: Exception on 18, channel 2 Jun 15 13:33:12
> DEBUG[11577] chan_zap.c: Got event Ring/Answered(2) on channel 2 (index 0)
> DEBUG[11577] chan_zap.c: Exception on 18, channel 2 Jun 15 13:33:20
> DEBUG[11577] chan_zap.c: Got event On hook(1) on channel 2 (index 0) Jun 15
> 13:33:20
> DEBUG[11577] channel.c: Didn't get a frame from channel: Zap/2-1 Jun 15
> 13:33:20
> DEBUG[11577] channel.c: Bridge stops bridging channels SIP/137-b1f2 and
> Zap/2-1 Jun 15 13:33:20
> DEBUG[11577] chan_zap.c: Hangup: channel: 2 index = 0, normal = 18, callwait
> = -1, thirdcall = -1 Jun 15 13:33:20 DEBUG[11577] chan_zap.c: Set option TDD
> MODE, value: OFF(0) on Zap/2-1 Jun 15 13:33:20
> DEBUG[11577] chan_zap.c: Updated conferencing on 2, with 0 conference users
>
>
> Thanks as always!
>
> _______________________________________________
> --Bandwidth and Colocation provided by Easynews.com --
>
> Asterisk-Users mailing list
> To UNSUBSCRIBE or update options visit:
>    http://lists.digium.com/mailman/listinfo/asterisk-users
>