Re: [hylafax-users] hylafax kills crontab and freezes solaris server

["Willem Oldeman" <willem.oldeman@home.nl>]

> We have hylafax-v4.0pl2, solaris 2.6, a multitech mt2834zdx modem, and a
US
> Robotics sportster 0459 modem.  About once a week, but on a completely
> random basis, our unix server freezes and we have to reboot.  Every time
> this happens there is either a very recent sendfax command or a
> communication with the modem (comm file in /var/spool/fax/log).  The first
> detectable symptom is that crontab stops submitting jobs. Slowly over an
> hour or two, other services start dying, such as databases, web servers,
> telnet etc.
>

It could be that some interrupt line is going crazy and eats up all
resources. I don't know if your machine has some way to view the number of
interrupts coming in, but on my Linux machine it goes like this:

$ cat /proc/interrupts
           CPU0
  0:  147039031          XT-PIC  timer
  1:          4          XT-PIC  keyboard
  2:          0          XT-PIC  cascade
  3:     225228          XT-PIC  serial
  7:      95622          XT-PIC  HiSax
 10:    2641063          XT-PIC  eth1
 11:   12102224          XT-PIC  eth0
 13:          1          XT-PIC  fpu
 14:    2339577          XT-PIC  ide0
NMI:          0
ERR:          0

You could check if your serial interrupt counter increments like crazy...
I'm not sure about this, it's a wild guess:) Also the 'top' command may be
handy to see what is eating CPU time/memory and makes your machine grind to
a halt.

Good luck,
Willem




____________________ HylaFAX(tm) Users Mailing List _______________________
 To unsub: mail -s unsubscribe hylafax-users-request@hylafax.org < /dev/null