I need a script that will reboot the server when "Unable to handle kernel NULL pointer dereference at virtual address" is found in the /var/log/messages file.

Something like this may work?

#!/usr/bin/perl -w

$output = `tail -f /var/log/messages`;
`/sbin/reboot` if $output =~ /Unable to handle kernel NULL pointer dereference at virtual address/;

--------------------------------

SSL Doesn't work when I setup an IP based domain with ssl enabled and a self signed certificate is installed. It always looks in the default directory for the server. (/usr/local/psa/home/vhosts/default/httpdocs/some-file-here) The other guy on the plesk forum that had this problem ended up having Plesk support rebuild apache to get SSL to work.

I think these kernel panics and the server not setup correct to support SSL is related.

A couple of weeks after I got the server and started to put members on it is when I noticed that ssl wasn't working. I did test almost everything when I first received this server because my 1st server was all messed up when I got it.... I failed to test ssl on my 2nd server. I think it was broke from the beginning but I cannot be sure. Too late now to even think that rackshack will look at it. I'll remember to test SSL next time.

Fixing it would be better, but it looks like Apache will need to rebuilt and I'll end up screwing something up even worst than what it is now. I'll just try to automatically reboot server when my -DAILY- "Unable to handle kernel NULL pointer dereference at virtual address" appears in my logs and crashes my server.

PS: Damn server crashed a few days after receive it from rackshack with the same error when I had done nothing to the server but installed pmfirewall, snort, ssh2 on it. I thought it was just a fluke. It goes down daily with this now.

Webmin might have what you need, check out this thread:
http://forum.rackshack.net/showthread.php?...&highlight=cron

Otherwise, you would probably be best off to write a cron job. Here is the basic syntax, sorry I didn't go further, but it's 5am.

if (expr) then
command
endif

Thanks!

I think this might have been fixed today. It has only been a few hours since it's been up so i'm not sure.

It was bad stick of memory.

7/31/02-4:29:32 PM - RackShack - Online - Reboot
7/31/02-7:48:40 PM - RackShack - Online - Reboot
7/31/02-9:05:50 PM - RackShack - RS - Test Ram
7/21/02-4:25:51 PM - RackShack - Online - Reboot
7/18/02-12:00:21 PM - RackShack - Online - Reboot
7/14/02-10:51:19 PM - RackShack - Online - Reboot
7/13/02-11:56:43 AM - RackShack - Online - Reboot
7/12/02-6:49:21 AM - RackShack - Online - Reboot
7/10/02-3:16:44 AM - RackShack - Online - Reboot
7/9/02-4:54:57 PM - RackShack - Online - Reboot
7/7/02-2:15:13 PM - RackShack - Online - Reboot
7/7/02-11:16:12 PM - RackShack - Online - Reboot
7/3/02-3:13:26 AM - RackShack - Online - Reboot
6/30/02-4:14:50 AM - RackShack - Online - Reboot
6/28/02-5:01:53 AM - RackShack - Online - Reboot
6/28/02-5:03:44 AM - RackShack - Online - Reboot
6/26/02-1:15:20 AM - RackShack - Online - Reboot
6/26/02-2:42:59 AM - RackShack - Online - Reboot
6/25/02-3:08:54 AM - RackShack - Online - Reboot
6/22/02-3:22:08 AM - RackShack - Online - Reboot
6/22/02-5:16:30 AM - RackShack - Online - Reboot
6/21/02-3:26:26 AM - RackShack - Online - Reboot
6/18/02-3:17:59 AM - RackShack - Online - Reboot
6/12/02-3:53:21 PM - RackShack - Online - Reboot

I got the same problems as you!
Did you fix it?

bad memory

check your logs when a crash occurs and if it looks like its for no reason have rackshack check your memory

--------------------------------------------

# uptime
5:07pm up 157 days, 8:11, 1 user, load average: 1.23, 1.37, 1.16








edit:
as for ssl, that was fixed when I had PLESK upgrade their software on the server since the upgrade would always fail when i tried to uprade that server.

The errors are Unable to handle kernel like you said above,
when this ocurred, the server definitely died.

maybe this is hardware's problems.

thank you for your help.

Suggestion::

run a software memtest before asking RS to investigate your ram....then include the results in your TT

###these instructions will set memtest to run 2 cycles and exit...there are many other options..you may want to read the documentation###


mkdir /home/admin/nh
cd /home/admin/nh
vi memtest.sh
###insert this############
#!/bin/bash
cd /home/admin/nh
wget http://www.qcc.sk.ca/~charlesc/software/me...r-2.93.1.tar.gz
tar xzf memtester-2.93.1.tar.gz
cd memtester-2.93.1
make
./memtest all 2 -l
####end of insert############
chmod 755 memtest.sh
sh memtest.sh


hope that helps