Reboot server nightly

[JohnnyBeGood]

So I need to schedule nightly reboot ie. 4am
I tried to schedule using webmin but it won't execute.
Any suggestions?

Thanks

 

[wardmundy]

http://pbxinaflash.com/community/th...reboot-plugin-script-for-piaf.1518/#post-8878

But why???

 

[JohnnyBeGood]

Thanks for taking time to reply!

For some odd reason while talking other party randomly mutes for 1-2 seconds. My server is P4 with 512MB of ram and its handling 1 call at a time. I'm not sure what's causing it.

 

[moyses]

I also facing the same problem! I have not been able to identify thee cause of this problem.

Searching in other forums some users suggest this could be a memory problem. They suggest doing the following to clean up the memory.

echo 3 > /proc/sys/vm/drop_caches
echo 2 > /proc/sys/vm/drop_caches
echo 1 > /proc/sys/vm/drop_caches

 

[therock112]

I find that I have to reboot my piaf 1.1 every 3 days or so.

what happens after 3 days you ask?

well my unlimitel and les.net trunks go down, I have attempted to do a amport restart blah blah blah and no help.

a complete restart appears to be the only cure and I got 512mb ram, pentium III 900mhz system.

I am willing to migrate to piaf 1.2 1GB ram with P4 1.6GHZ, not sure throwing more hardware is going to solve this issue.

this has been ongoing since I moved to piaf 1.1 back in January from a@h

 

[wardmundy]

We run our $199 WalMart specials for months at a time without a reboot. So... migrating to more current hardware may definitely help.

 

[bgriff]

Ward,

Don't sell 1.1 short, I have it running on a Compaq EN Deskpro with a 933 mHZ PIII and 512MB of ram and it is working great for me. The box has a digium card, 3 sip providers and IAX trunks to a similiar box. The same hardware even ran the "distro named after a disease 2.2" pretty well.

Thanks to the PiaF and FreePBX Teams for the great works!

 

[wardmundy]

Actually, our home systems still are running 1.1, too. we hope to have a one-way upgrade path to 1.2 very soon. Stay tuned!

 

[JohnnyBeGood]

Thanks for the replies!

I added echo 3 > /proc/sys/vm/drop_caches
to my cron job, I'm hopping its gonna help http://www.linuxinsight.com/proc_sys_vm_drop_caches.html

 

[thunderheart]

Upgrade! Oh my lord

Cool! You and the rest of the PIAF team are my heroes! I REALLY didn't want to have to reformat in order to upgrade to 1.2 so if you can come up with an upgrade path that avoids that, I for one would sure appreciate it.

Dallas

 

[stuck]

Any updates on this? is the cache flushing helping your situation?

 

[therock112]

Unfortunately every 2 to 3 days I still have to reboot the * box....

 

[jmullinix]

What version of Asterisk are you running? There was one about the time of 1.1 that had a huge memory leak.

 

[therock112]

I am running piaf 1.3
Centos 5.2
asterisk 1.4.21.2

iptables turned off

this is not a busy box, about 7 sip extension, 2 iax trunks, 2 sip trunks, 1 zap ...nothing fancy.

every 3 to 4 days, the iax and sip trunks will not register, restarting amp doesnt help, a full reboot does.

 

[Bitnetix]

I'm not condoning this, but if you REALLY want to reboot every day at (say) 3am local time, the simplest way to do this is:

echo "0 3 * * * /sbin/shutdown -t 30 -r +10" >> /etc/crontab

And you will start rebooting every day at 3am. The +10 says to give a ten minute warning, so this really does the reboot starting at 3:10am. The -t 30 tells processes "time to stop," (at 3:10) and then waits 30 seconds before killing any that didn't shut themselves down. You can change either of these numbers to suit your fancy, and the "0 3" at the beginning is the minutes and hour, respectively, that this command kicks off. Feel free to change any/all of this.

Most brute force would be to simply go at 3am with no warnings:

echo "0 3 * * * /sbin/shutdown -r now" >> /etc/crontab

 

[therock112]

thanks Bitnetix, this should do the "dirty" for now.

i would like to get to the root of this issue, but there is only so many hours in the day to do it all somehow.....

 

[Bitnetix]

Dirty is sometimes the best.

 

[stucker]

SIP Trunk Monitor

Anyone have a working script (php/sh/agi) to monitor failed SIP trunks? I'd like the script to be able to email along with reset the trunk or server if the trunk failed.

Seaching the forums (PiaF & Green) was not able to find a working script to place in the trunk configuration to monitor.

Any help or direction would be most appreciated. Thanks!

 

[therock112]

would this be of any help:

http://pbxinaflash.com/community/threads/email-alert-on-trunk-failure.1968/

 

[dad311]

This sounds like hardware issue. Have you tried replacing the NIC card? Maybe the NIC is using the wrong driver or has a hardware issue that only a reboot can fix.