[Asterisk-Users] SpanDSP + Lexmark 6170 = Cut off faxes?

[Michael Welter]

Scott Lykens wrote:
> On Tue, 09 Nov 2004 07:41:36 +0800, Steve Underwood <steveu at coppice.org> wrote:
> 
> 
>>I have looked at many similar reports, and all but one turned out to be
>>due to data slips. However, if you are sure this is only happening with
>>one particular FAX machine your problem might be different. In t30.c
>>uncomment the first line. Rebuild spandsp and try FAXing. You should get
>>audio log files in /tmp. Send them to me, and I will investigate.
> 
> 
> Steve:
> 
> It appears I shouldn't have spoken so soon, I apologize; that is,
> without a larger sample of faxes to observe. I was concerned about
> this particular machine because it fails on every page but in
> reviewing a sample of about 45 faxes this morning it appears that I am
> having trouble with other machines as well but not to the same extent
> I have with the Lexmark. The Lexmark fails on *every* page while some
> others fail only on one or two pages out of a fax, but still most
> others work fine.
> 
> For what its worth it also looks like the failures occur more
> frequently on faxes from longer distances away, I'm in Pennsylvania
> and the Lexmark is in Florida. It also appears that a machine in
> California is having an occasional problem with me too.
> 
> Before wasting any more of your time tracking down a problem that
> likely doesn't exist in the software I will approach it as a timing
> problem first.
> 
> On that tack, what should I look for as a potential source of a timing
> problem? As I indicated previously I see this problem even when I am
> connected directly to the PSTN and (believe) that I am using the PSTN
> as a clock source on the PRI.
> 
> Assuming that the clock from the telco is good should I start looking
> at the motherboard/system itself as a source of a timing problem? I
> have verified that the T1 card is on its own interrupt. Perhaps the
> motherboard just isn't cut out for handling this volume of interrupts?
> 
> Thank you for the help.
> 
> sl
> _______________________________________________
> Asterisk-Users mailing list
> Asterisk-Users at lists.digium.com
> http://lists.digium.com/mailman/listinfo/asterisk-users
> To UNSUBSCRIBE or update options visit:
>    http://lists.digium.com/mailman/listinfo/asterisk-users
> 
> 

Do you have a telco test number in your area that gives you silence?  If 
so, just call that number and listen carefully.  If not, loop-back a 
call on your T-1.

If you hear an occasional click or pop then you're hearing a frame slip 
on the T-1.  This is step one for diagnosing a fax problem.

I have a customer's P3TSSA MB (Pentium III) sitting here with a PCI SCSI 
controller.  With a T100P, I've tried every combination of PCI card 
placement and irq assignment, and I still get clicking and popping on 
the T-1.  Needless to say, fax transmission using spandsp is impossible.

I also have a GA-7N400 Pro2 MB (AMD XP 2800+) that I'm building for a 
customer.  It has the same Adtran 750 on the same T100P card as above. 
When I call from phone to phone there is no clicking or popping on the 
line.  Fax transmission using spandsp is beautiful on this system.

I would also recommend running top or system monitor while you're 
testing.  If your processor ever pegs at 100% then your fax transmission 
will probably fail.

Mike


-- 
Michael Welter
Introspect Telephony Corp.
Denver, Colorado US
+1.303.674.2575
mike at introspect.com
www.introspect.com