is really trixbox ready for a production environment ?
Submitted by drcnet on Tue, 09/04/2007 - 9:28am.
I've started to test Trixbox intensivly about six month ago (from 2.0 version).
I've installed it on many different machine (HP Proliant ML110, AMD sempron machine etc..., with different MB, memory, hard disk, networks card, video card, etc...)
Currently my trixbox 2.2 run on a AMD athlon 1.7 GHz machine and on it is installed a digium TDM400 analog card with one FXO module (to connect to PSTN).
Connected to my Trixbox server there are a lot of Grandstream phones (BT100, BT200, GXP 2000)
But on all different hardware configuration the problem is the same:
all works fine for about three-four weeks but beyond this period the server is "freezed" and all my voice network is down, I can't access trixbox server via his IP address (by browser) o by other means and the only way to fix thix(big..) problem is to phisycally go to the server, press the power key and restart the machine.
By this way all works fine until the next stop (three o four weeks later...).
I'm very frustrated for this.
I don't know why this.
Using different hardware don't fix the problem.
Any ideas?
Thanks in advance.
Cosimo MERCURO
Our oldest system at a
Tue, 09/04/2007 - 9:31am
kernel panic
Tue, 09/04/2007 - 9:41am
I don't remember exactly what is on the screen.
May be Kernel panic.
But the reason of my post is to learn a way to understand (and fix...) what happen on my servers.
I work from about seven year on Windwos Server System (2000-2003) and all that happen is write on Event Viewer.
This is very useful in troubeshooting process.
Is there an equivalent tool for linux systems?
Cosimo
devil is in the detail.
Tue, 09/04/2007 - 11:44pm
System Resources . . . .
Wed, 09/05/2007 - 1:53am
OK ! I had a similar problem with one of my production servers. The only way I fixed this was by upgrading my hardware to a better one.
There is nothing wrong with the software or the hardware itself. There is a very genuine and logical solution to the problem (Atleast it worked for me)
A system would freeze most of the time when it would have more load than it could handle. The system resources might not be adequate to manage the work load. I had a production server with a2billing set up on it. The server used to freeze after every 5-6 days. Consequently, I had to set up my billing on a different server. Now the calls are originated/terminated through one server and billed on a different server. Everything seems to work fine since then.
How many concurrent calls do you have ?
To answer your question, yes
Wed, 09/05/2007 - 6:16pm
To answer your question, yes there is an event log of sorts in /var/log/asterisk. You can also look at the last 2000 lines of the log file from the FreePBX GUI. Lock ups/crashes are easy because all you have to do is look at the last few things before the lock up. The problem is not always obvious from the info in there but there is a good chance it is.
If it looks like a bug submit it to bugs.digium.com AFTER you do a search of other bug reports to see if there is a similar issue.
Member Since:
2006-09-10