PIAF server going down too often (5th time)

[sims]

I am using Everex m/c. i have posted about this before but we could not identify any reason and I decided to wait for more such instances.

Now it has happened at least 5 times. So it would be great if somebody could advise me on what to look for in which logs.

Also, if there is any external mechanism by which I could monitor my server being up or not, that is very needed at this point.

---

here are logs that i could pull up from /var/log/asterisk/full. the portion where it appears the system went down.

[Sep 20 14:16:57] NOTICE[29572] chan_sip.c: -- Registration for 'smalani@inb$
[Sep 20 14:17:17] NOTICE[29572] chan_sip.c: -- Registration for 'smalani@inb$
[Sep 20 14:19:55] NOTICE[29572] chan_sip.c: Peer '101' is now UNREACHABLE! Las$
[Sep 20 14:29:45] VERBOSE[29572] logger.c: -- Registered SIP '101' at 122.1$
[Sep 20 14:29:46] NOTICE[29572] chan_sip.c: Peer '101' is now Reachable. (405ms$
[Sep 20 14:29:46] VERBOSE[29572] logger.c: -- Registered SIP '1001' at 122.$
[Sep 21 10:18:41] VERBOSE[3055] logger.c: Asterisk Event Logger Started /var/lo$
[Sep 21 10:18:41] VERBOSE[3055] logger.c: Asterisk Dynamic Loader Starting:
[Sep 21 10:18:41] VERBOSE[3055] logger.c: == Parsing '/etc/asterisk/modules.c$
[Sep 21 10:18:41] VERBOSE[3055] logger.c: == Manager registered action Ping
[Sep 21 10:18:41] VERBOSE[3055] logger.c: == Manager registered action Events
[Sep 21 10:18:41] VERBOSE[3055] logger.c: == Manager registered action Logoff
[Sep 21 10:18:41] VERBOSE[3055] logger.c: == Manager registered action Hangup

 

[wardmundy]

Have you added any hardware to the machine?

If so, remove it and see if you still have the problem.

If not, the first thing I'd try is to load Tom King's full system backup utility and make a full backup to a flash drive. Then make CDs from the ISO's of the backup. Then reboot using the first CD. Restore your entire system which also will do a low level format on the hard disk. This cured some problems on one of our machines.

If this doesn't fix it, I would first replace the RAM and see if that fixes it.

If this doesn't fix it, I would reinstall the Orgasmatron build and start over after making a FreePBX backup and copying it off the machine so that it can be restored after you finish the reload.

 

[sims]

i have not added any hardware. will make a full backup as soon as i can get to it. and then do the restore as you have advised.

meanwhile, is there any way in which i can monitor my system so that at least i get to know when it is down.

thanks!

 

[wardmundy]

Sure. Add a line like this to /etc/sysconfig/iptables above the COMMIT line using the actual IP address of your server:

-A INPUT -d 192.168.0.50 -p icmp -j ACCEPT

Then...

service iptables stop
service iptables start

After that, you can ping the server from anywhere on your LAN.

 

[tel0p]

options

There are quite of few different choices for monitoring over here: http://www.voip-info.org/wiki/view/Asterisk+monitoring ...
'ping'ing the box won't actually tell you that the asterisk server is running or not.

 

[wardmundy]

The topic of the thread was PIAF server going down. I took that to mean the entire server was crashing, not just Asterisk. Pings would detect that, but the suggestions for monitoring Asterisk are excellent as well, just a bit more complicated.

 

[tel0p]

ahh

My bad, I thought his asterisk process was crashing.. What I use to monitor my server(s) is montastic.com . Get a txt whenever a box goes down or up. Works well.

As long as there's an HTTP port open to the world you're good to go.

 

[powerontech]

There is a very simple mechanism for getting notified of problems without installing any additional software. It is explained near the bottom in the "Getting notified of problems" section.

http://www.powerpbx.org/content/freepbx-asterisk-and-linux-advanced-topics