[Asterisk-Users] Anyone using * in a live production environment?

[Steven Critchfield]

On Wed, 2003-11-05 at 03:11, Gavin Hamill wrote:
> On Tue, Nov 04, 2003 at 01:52:46PM -0600, Steven Critchfield wrote:
> 
> > - Has been in nearly fault free operation for more than since 05-2002.
> 
> Great stuff, Steven! :)
> 
> Can I enquire what was the cause of the downtime? Was it planned-
> maintenance, or an actual fault with the Asterisk software / Digium
> hardware?

One failure was a kernel lockup when compiling a new module to start
testing ZapRas.
I have made a couple of stupid mistakes that shut asterisk down.
Outside of the above comments, my gateway machine has worked flawlessly,
but this is also why it isn't given much to do. It is too important to
have something make it fail.

On my pbx machine in the office though, we occasionally have small
failures. Most recently we had a segfault show up in a zapata handle
event function, but couldn't track it down well enough to report upon
it. This machine is specifically set up to be more of a test bed
machine. We have only 4 people currently in our office, and 2 of us use
the phones mainly for testing of our software.  

> Roughly how long has the system been down in total since 'going live' ?

I'd say we haven't had more than 10 minutes downtime on our gateway
machine, and thats mostly due to the kernel lockup that caused me to
have to call my colo facility to do a hands on reset of the machine.

We may have about that much time on our pbx, but this is also where we
test our patches to asterisk, so it can't be called against asterisk.

My execution of this setup hasn't been telco quality, but seems pretty
on par with small office pbx systems.

-- 
Steven Critchfield <critch at basesys.com>