Re: [hylafax-users] MultiTech MT5634ZPX-PCI-V92 and Wedging

["Christian M. Watts" <cmwatts@xxxxxxxxxxxxx>]

>
> Top posting moved to bottom:
>>
>> On 2005.02.18 12:50 Steve Tuckner wrote:
>>> Not being a hylafax expert it makes me wonder about how that call
>>> ended.
>>> There was a problem earlier with the ZPX-PCI and Hylafax where Hylafax
>>> would
>>> drop DTR instead of hanging up with an ATH. Could this be the issue
>>> here?
>>>
>>> Steve Tuckner
>>> Multi-Tech Systems
>>>
>>> > -----Original Message-----
>>> > From: hylafax-users-bounce@xxxxxxxxxxx
>>> > [mailto:hylafax-users-bounce@xxxxxxxxxxx]On Behalf Of Mike McMullen
>>> > Sent: Friday, February 18, 2005 2:11 PM
>>> > To: hylafax-users@xxxxxxxxxxx
>>> > Subject: Re: [hylafax-users] MultiTech MT5634ZPX-PCI-V92 and Wedging
>>> >
>>> >
>>> >
>>> > ----- Original Message -----
>>> > From: "Lee Howard" <faxguy@xxxxxxxxxxxxxxxx>
>>> > To: "Mike McMullen" <mlm@xxxxxxxxxxxxxxxxxx>
>>> > Cc: <hylafax-users@xxxxxxxxxxx>
>>> > Sent: Wednesday, February 16, 2005 9:50 AM
>>> > Subject: Re: [hylafax-users] MultiTech MT5634ZPX-PCI-V92 and Wedging
>>> >
>>> >
>>> > > On 2005.02.16 09:35 Mike McMullen wrote:
>>> > >
>>> > >> I have noticed that right before the wedging occurs, a fax was
>>> > >> received
>>> > >> that failed with the following message:
>>> > >
>>> > >>    Received premature V.34 termination.
>>> > >>
>>> > >> I'm also seeing as I posted yesterday:
>>> > >>
>>> > >> "Failed to properly detect hight-speed data carrier"
>>> > >>
>>> > >> And sometimes the modem wedges after that.
>>> > >>
>>> > >> I've attached the config file for one of the modems. Wedging
>>> errors
>>> > >> appear with equal frequency on either modem.
>>> > >>
>>> > >> I can supply logs if someone thinks that would help.
>>> > >>
>>> > >> Any insight appreciated!
>>> > >
>>> > > The thing to do would be to see if the modem really is
>>> > wedged.  So when
>>> > > the modem wedges, without power-cycling the modem or
>>> > otherwise changing
>>> > > things, run 'minicom -o' on it and see if you can get it to
>>> respond
>>> > > after a few ATZs or something.  If that doesn't work, then
>>> > maybe let
>>> > > minicom initialize it (by omitting the -o).  If you can get it to
>>> > > respond, then HylaFAX needs some improvement with the
>>> > "reset" routine.
>>> > > If you can't, then the modem really is hung, and it's
>>> > likely a firmware
>>> > > issue that MultiTech will need to examine.
>>> > >
>>> > > The session log that causes the modem to hang/wedge would
>>> > be useful in
>>> > > assisting you.
>>> > >
>>> > > Lee.
>>> >
>>> > Hi Lee/Bill and all other interested parties,
>>> >
>>> > One of the modems wedged this morning and per your
>>> > suggestion,  I turned off
>>> > the faxgetty on the modem and removed it from hylafax and ran
>>> minicom
>>> > on it. Using the speed in the conf.modem file of 56K I was
>>> > unable to get
>>> > any response at all from the modem.
>>> >
>>> > I ran it at 38.4K just to cover the bases and no response at
>>> > all. So I had to remove the
>>> > phone lines, reboot and everything was back to normal.
>>> >
>>> > An incoming fax attempt failed about 5 times before the modem
>>> > wedged. After the
>>> > last failed attempt it appears the modem wedged. I've
>>> > attached the com log if that
>>> > can provide any help.
>>> >
>>> > Any insight appreciated!
>>> >
>>> > Mike
>>> >
>>> >
>
>> Steve,
>>
>> On both SuperG3 and non-SuperG3 calls HylaFAX sends ATH0 before dropping
>> DTR.  Mike would need to
>> examine his server logs to make sure that it happened in that particular
>> instance, but that's what
>> it's *supposed* to do in 4.2.1, at least.
>>
>> However, if for some reason ATH0 results with ERROR or no response at
>> all, then it will just move
>> on after 5 seconds.
>>
>> Lee.
>
> Ok. Different modem wedged this time. For all trying to keep up I have two
> ZPX-PCIs. I set server to 0xFFF. A fax came in and got a timeout and
> failed.
> Right after that modem1 wedged. While the com log didn't show anything,
> looking
> at /var/log/messages showed some errors.
>
> First I got a kernel error "LSR safety check engaged!" before FaxGetty
> reports
> "MODEM setBaudRate::tcgetattr: Input/output error".
>
> I assume that is the root of the problem but I have no idea what the
> kernel error
> is stating.
>
> The trace is below. This is a FC2 system running on an a 700 MHZ Athlon
> circa
> 2000. Thats when I bought it.
>
> Mike
>


The "LSR safety check" kernel error means that the serial driver thinks
something 'bad' happened with the port that it was talking to. My
understanding is that the UART returned something invalid when it wasn't
supposed to. When this happens, the kernel gives up on the port and you're
pretty well dead in the water.

Some documentation is in the Modem HOWTO at (this specific error is
mentioned):

http://www.tldp.org/HOWTO/Modem-HOWTO.php

You might also search the hylafax archives, as I believe Lee has fielded
this issue before (earlier last year):

http://www.hylafax.org/archive/2004-04/msg00040.php


Bottom line is that something flaky is going on between the kernel serial
driver and your modem ... possibly IRQ sharing, I/O conflict or something
isn't plugged in quite right, that type of thing.

Christian

____________________ HylaFAX(tm) Users Mailing List _______________________
  To subscribe/unsubscribe, click http://lists.hylafax.org/cgi-bin/lsg2.cgi
 On UNIX: mail -s unsubscribe hylafax-users-request@xxxxxxxxxxx < /dev/null
  *To learn about commercial HylaFAX(tm) support, mail sales@xxxxxxxxx*