Freepbx server freezing every hour or so

This just started happening two days ago… When I updated the queues and paging modules I think… I am completely up to date with the latest greatest modules… Is there some cron job that is running every hour that is chewing up all the processing power? If yes, is there some way I can turn it off?

I am having to reboot the server when this happens as it freezes and does not return on its own for at least 20min or so if not rebooted.

The system has worked beautifully for two years now, with various versions of freepbx… This 2.3.0.x release is the first time I have had any issues. Since this is our primary phone system, everytime this happens, all calls are lost…

Any help would be appreciated.

What do you meen by “freezing” Does the server lock up? Does Asterisk restart?
A little more info…

Bill

there is a cron job that runs every hour, owned by cron. It runs the ‘cron_manager’ in FreePBX that checks if it needs to look for updates and in the future, will check for anything else that may be scheduled. This was added in 2.3 and not present before but has been in 2.3 since the middle of the beta program.

If you suspect there may be an issue going on wrt to that then you could manually remove the cron process to see if it is the cause. (And if so, it would be interesting to now why). The cron job runs as user asterisk so as root a “crontab -eu asterisk” should allow you to edit the job (you may see backup jobs also, leave those, they have not been moved into the cronmanager yet). At this point the only thing that will effect are the periodic checks for online updates, which should be running once a day at most. (And turning that off will not stop the cronmanager since it will still wake up and check if any jobs need to be run). You could try disabling the online update checking first though before removing the cron entry to see if it makes a difference.

Also - I can’t recall now, but it may be that the next time you do any change to the system where you press the orange bar, it reinstalls the cronjob:( - so take a look and check on that.

If it is the cronjob, the real question is why? It can be run manually as well in case it gives any hints. I have not heard any other feedback similar to what you are experiencing.

Philippe Lindheimer - FreePBX Project Lead
http//freepbx.org - IRC #freepbx

Yes, the server locks up - asterisk does not seem to be restarting - just the server locks up and all calls stop, and 20min later the server is fine again. Again, this just started two days ago. Before that even 2.3.0.x has been working like a charm.

Will try the turning off of the update checks…

I just had something else wierd happen. I was on a call, that just dropped half way through (was a tollfree call going through an ENUM trunk) and now the call will not connect at all (the logs just show a CDR insertion saying congestion)

If you can get on the machine when it’s acting up, you can type ‘top’ in the linux prompt, it will show you all running processes. Then you can press ‘P’ to sort by CPU usage or ‘M’ to sort by memory usage. This will tell you if it’s something running that’s hogging resources.

This is still happening - even after turning off the updates - though less frequently now. Unfortunately, when this happens, the console does not respond if already connected, and if not connected, I am unable to connect…

well… it could be a network glitch. Can you ping the server during this time? Don’t rule out hardware.
It could be that the NIC, switch, or router is locking up. Check uptime - is it rebooting?
netstat -i and look for errors.
Also, do a dmesg | grep eth to see any ethernet disconnect related messages.

netstat -i reports NO errors
Uptime shows that the system is NOT rebooting when it locks up - the system has been up for 12 days after it comes back…
dmesg | grep eth also does not show any issues
So unlikely to be a network issue - I think it might be CPU going out of control during that time…

Is there a log for the cron jobs that run? So I can check when they ran and for how long?

Check the log files/var/log and /var/log/asterisk and see what it says around the time of the reboots.

If it just started happening with no apparent changes from you and it’s the whole server including CentOS locking up then look into hardware. What motherboard are you using? Fans working? Did you blow any excessive dust out of it?