[hylafax-users] R: R: R: Hylafax Freeze with no error or warning

["Giuseppe Dia" <giuseppe.dia@xxxxxxxxxx>]

Da: hylafax-users-bounce@xxxxxxxxxxx
[mailto:hylafax-users-bounce@xxxxxxxxxxx] Per conto di Aidan Van Dyk
Inviato: giovedì 23 dicembre 2010 15.13
A: Giuseppe Dia
Cc: Lee Howard; hylafax-users@xxxxxxxxxxx
Oggetto: Re: [hylafax-users] R: R: Hylafax Freeze with no error or warning

On Thu, Dec 23, 2010 at 3:54 AM, Giuseppe Dia <giuseppe.dia@xxxxxxxxxx>
wrote:

> Seems a known bug here:
> https://bugs.launchpad.net/ubuntu/+source/hylafax/+bug/600219
> I think I should stick in a watchdog cron script, as David Forrest
> suggested. My mileage will vary indeed as I'm forced to restart hylafax
> altogheter rather than spawning new faxgetty, because having many of them
I
> don't really know which one fails (I must say they all fail, to be fair).
> What's your take on this?
> Bye and thanks for the help so far.

>If that's the same bug, and you can reproduce and have basic skills at
>GDB, or linux, we'ld love to get a copy of a core file, or back trace
>or anything really on the crashing faxgetty.


Thanks for your answer. If it does it again I will provide you with the core
files. At the moment I had to solve things quickly for obvious reasons and I
had no more crashes.


>None of the "backtraces" or dump in that ticket are actually faxgetty,
>they are all lots of various children of faxgetty (like tiff2pdf,

This remind me it stated after I enabled PDF as filetype.



>called from faxrcvd, during a non-crashing faxgetty run).  And both
>reported it "fixed" after upgrading kernel and replacing faulty
>hardware...

I followed the bugreport advice of upgrading the kernel from
linux-image-2.6.35-22-generic-pae on x86 (last one on ubuntu 10.10).
This only made things worse, with faxgettys started dying after successful
reception of  faxes minute by minute. Before the upgrade they use to die
once in a week. I then followed the other advice in the bugreport: I
disabled apparmor and setup a cron job to check for the faxgetty every
minute or so and send a warning when one of the faxgetty (say faxgetty -D
ttyACM0) would die.
So far I'm seeing all of them, no warning, and everything is just fine. Too
soon to say if this solved the problem, but at least they stopped dying
minute by minute. I'll keep you posted with the result and send you the core
files, if it happens to die again.
I will also try taking out Filetype: pdf.
Thanks,
GD



____________________ 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*