[asterisk-users] What do you do to keep asterisk alive?

[Per Jessen]

Matthew J. Roth wrote:

> Per Jessen wrote:
>> I don't know why it's stopping, but I'm pretty certain it's a
>> segfault. Next time it happens, I should be getting the core dump.
>> I'm running 1.4.13, no AGI scripts.
> Per,
> 
> You should be able to determine if it was a segfault by looking at
> your system log.  For example, on one of my CentOS systems:
> 
>   [root at ast01 ~]# grep segfault /var/log/messages
>   Jul 31 11:10:41 ast01 kernel: asterisk[11548]: segfault at
> 0000000000000010 rip 0000000000420450 rsp 0000000041c670e0 error 4

Hi Matthew

I've checked my logs, and there's nothing like that - but I don't think
I've ever seen a logline being written because of a segfault.  I wonder
if it's a special config option. 

> Hopefully, you don't have enough segfaults occurring on your system to
> require a more complex search.  

Nah, right now asterisk is the only service that is a little unstable. 

> I also recommend the safe_asterisk 
> script if your failures are all segfaults.  Unfortunately, the other
> way Asterisk dies (the process is still alive, but not really doing
> anything) isn't as easy to diagnose.

an strace might reveal what it's doing - or where. 



/Per Jessen, Zürich

-- 
http://www.spamchek.com/ - your spam is our business.