server crash every 2nd day @ same time :)

[theBassTender]

I'm still experiencing this issue as well - removed laus from the system - did not fix the issue.

I'm running a single pentium 4 with sata raid with RedHat Enterprise Linux 3 and cpanel 10.0.0R7. Anybody find any other solutions?

[jeroman8]

Well, the issue was resolved for me with kernel 2.6.
Don't know which RAID controller I have but I guess it's not all controllers
having the issue.

[Hostbox.be]

I have the same problem with my new server since last week, every other day the server hangs after /scripts/upcp in CRON. Only ping does still work and only a power reboot brings the server back online.

I have CentOS 3.4 installed with kernel 2.4.21-27.0.1
Latest stable release of cpanel: WHM 10.0.0 cPanel 10.0.0-R7
It's a servelinux server L4 with two IDE hard drives in RAID 1

FYI: I upgraded my perl version just after the fresh installation to:
This is perl, v5.8.4 built for i686-linux
perl-5.8.0-88.9 (rpm version)


So now every 2 days, I have to reboot my server at night around 4am, just after /scripts/upcp ?

What is going on because I am not the single user who has this problem ?
And I cannot see what's on the screen when the server crashes, because the server is 2 hours driving from my home.

What can I do ?

[hostcp3]

We had a brand new server do this also, had DC change the Ram and has not crashed since.

It would die everyday at same time running the cron cpanel update

a manual update would not kill it for some reason, replaced ram and all fixed.

[Hostbox.be]

I changed time for /scripts/upcp in CRON to 23h45, far away from /cpbackup at 01h00, and the first night upcp went well. BUT I saw at that time the yum process was using almost 90 to 95% of my CPU during a couple of minutes ? Is this normal ?

Last night my server crashed again a little after 23h45, so back the upcp problem :-(

Could it be yum that is eating up al the memory and using the whole CPU that causes the problem ? Is there a solution for this ?

[panayot]

Had same problem but server died every day during upcp. Problem got fixed using bbailey's solution. Thanks for that! Read above his full explanation or I here post just the necessary steps:

root@localhost# chkconfig audit off
root@localhost# service audit stop
root@localhost# ps -ef | grep auditd (make sure it's stopped)

It also involves a kernel module (named "audit"), which you may also want to disable. Doing so will prevent the userspace tools that support auditing from generating errors when they can no longer find /dev/audit.

root@localhost# service crond stop
root@localhost# service atd stop
root@localhost# rmmod audit
root@localhost# lsmod | grep audit (make sure it's gone)
root@localhost# echo "alias char-major-10-224 off" >> /etc/modules.conf
root@localhost# service crond start
root@localhost# service atd start

You could even go so far as to remove the laus package altogether.

root@localhost# rpm -e laus

[Hostbox.be]

And this laus package is not needed by CentOS or Cpanel ?
What does audit do, because I don't know what it is ?

So I can stop en remove this functions without causing other problems ?

[kosmo]

One of my servers started crashing some weeks ago, once or so every week, then it got more and more until it was daily at 6 in the morning. After I removed
0 6 * * * /scripts/exim_tidydb > /dev/null 2>&1
from the root cronjob it works fine (until now).

crash = it got loads like 700+ and no service responded.

Will the removal of this cronjob hurt?

kosmo

[icemanprime]

We had a brand new server do this also, had DC change the Ram and has not crashed since.

It would die everyday at same time running the cron cpanel update

a manual update would not kill it for some reason, replaced ram and all fixed.

I was experiencing a hang (only ping would respond) every Sunday but it would run fine all week... replacing the RAM solved the issue.

[SloanPeterson]

I get these errors:
root@server100 [~]# chkconfig audit off
error reading information on service audit: No such fil
root@server100 [~]# service audit stop
audit: unrecognized service

Why won't that work?

[uniondatacom]

Had this exact same problem on a Dell SC420 with RH3 ES, I just reinstalled the server from scratch and found this thread after hehe, but I got rid of audit just in case, I dont want it crashing again. Anyway, I have some more information that I dont see anyone has posted here, the server actually crashes with a kernel panic (you will notice the keyboard lights flashing once its hung up) and it throws an error message to the console. It is also supposed to save this to /var/log/messages but I guess its a really severe crash so it doesnt.

Anyway, if youre familiar with kernel panic, this trace thrown to console shows exactly what kernel module caused the crash, BUT I have not been able to get the full panic message since it scrolls way beyond the console screen

Also, the message is not thrown to the console unless you log in and type this:
* setterm -blank 0
* setterm -powerdown 0
* setterm -powersave off

The little information that I do have from the panic message is this:

Call Trace: [<f8a02718>]linux_sysent[audit] 0x338 (0xca519d04)

more audit crap....

Code: 8b 9c 83 00 00 00 c0 c7 04 24 68 df 28 c0 89 5c 24 04 e8 ae

Kernel Panic: Fatal Exception

So I guess audit was the culprit, but having more of the panic message would help a lot more.

I realize that most of you probably dont have access to your actual server since theyre either colocated or dedicated servers, but I am an admin for such a center so I do, and I hope this helps everyone, and helps cPanel too.

[mum]

To SloanPeterson .

Try auditd instead of audit like :

chkconfig auditd off

[dreamwiz]

Looks like audit problem. You should check http://www.eth0.us/auditd for more information.

[chican0]

I am on CentOS as well.

WHM 10.8.0 cPanel 10.8.1-R95
CentOS 4.3 i686 - WHM X v3.1.0

Below is a copy of what I have done...

root@pya01 [/var/log]# ps -ef | grep auditd
root 1238 13 0 22:45 ? 00:00:00 [kauditd]
root 8180 6460 0 23:00 pts/0 00:00:00 grep auditd
root@pya01 [/var/log]# chkconfig audit off
error reading information on service audit: No such file or directory
root@pya01 [/var/log]# chkconfig auditd off
root@pya01 [/var/log]# service audit stop
audit: unrecognized service
root@pya01 [/var/log]# service auditd stop
Stopping auditd: [FAILED]
root@pya01 [/var/log]# ps -ef | grep auditd
root 1238 13 0 22:45 ? 00:00:00 [kauditd]
root 8262 6460 0 23:00 pts/0 00:00:00 grep auditd
root@pya01 [/var/log]#

Would that mean that my issue may be resolved now? Is kauditd the same thing as auditd?

[neilio]

Maybe I'm a total idiot (it's possible!) but when I run "chkconfig auditd off" I get the error "chkconfig: command not found".

This is on a RHE 3.0 machine. Any ideas?