Hi all, Recently one my Centos 5.7 VM just crashes at least once a day randomly (hang).
In /var/log/messages there is nothing at all that there is problem (no error, no failure). The log just stops.
The only change I did before this crashes is I activated LDAP authentication, and also auditd. But I don't see any evidence relating to it.
Any clue where to look for the cause?
Thank you. Fajar
Vreme: 11/10/2011 10:03 AM, Fajar Priyanto piše:
Hi all, Recently one my Centos 5.7 VM just crashes at least once a day randomly (hang).
In /var/log/messages there is nothing at all that there is problem (no error, no failure). The log just stops.
The only change I did before this crashes is I activated LDAP authentication, and also auditd. But I don't see any evidence relating to it.
Any clue where to look for the cause?
Best course should be to disable those apps and try without them if possible.
Beside that, first find out exact time of the crash (is there any cron job associated) and check your file systems for error. Also check is you have enough free space.
Fajar Priyanto wrote:
Hi all, Recently one my Centos 5.7 VM just crashes at least once a day randomly (hang).
In /var/log/messages there is nothing at all that there is problem (no error, no failure). The log just stops.
The only change I did before this crashes is I activated LDAP authentication, and also auditd. But I don't see any evidence relating to it.
Any clue where to look for the cause?
Two questions: is there anything on the console screen? Does it just hang, or reboot?
If the latter, and there's nothing in the logs, it's possible activating openLDAP was just coincidental with the problem, and I'd start worrying about hardware problems.
mark, waiting on an FE who's an hour late
On Thu, Nov 10, 2011 at 10:25 PM, m.roth@5-cent.us wrote:
Any clue where to look for the cause?
Two questions: is there anything on the console screen? Does it just hang, or reboot?
If the latter, and there's nothing in the logs, it's possible activating openLDAP was just coincidental with the problem, and I'd start worrying about hardware problems.
1. It's a VM, the console looks black. Typing something on keyboard doesn't bring back the console to alive. It's also unresponsive to ping.
2. It's just hang. Not reboot. I have to power off the VM and power on. The ESXi host has many VM in it and only my that VM has problem.
No specific time of hang.
Only thing I can try is to deactivate auditd. Let's see if it survives more than one day without hang.
On Nov 10, 2011, at 9:59 PM, Fajar Priyanto fajarpri@arinet.org wrote:
On Thu, Nov 10, 2011 at 10:25 PM, m.roth@5-cent.us wrote:
Any clue where to look for the cause?
Two questions: is there anything on the console screen? Does it just hang, or reboot?
If the latter, and there's nothing in the logs, it's possible activating openLDAP was just coincidental with the problem, and I'd start worrying about hardware problems.
- It's a VM, the console looks black. Typing something on keyboard
doesn't bring back the console to alive. It's also unresponsive to ping.
- It's just hang. Not reboot. I have to power off the VM and power on.
The ESXi host has many VM in it and only my that VM has problem.
No specific time of hang.
Only thing I can try is to deactivate auditd. Let's see if it survives more than one day without hang. _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
On Nov 10, 2011, at 9:59 PM, Fajar Priyanto fajarpri@arinet.org wrote:
On Thu, Nov 10, 2011 at 10:25 PM, m.roth@5-cent.us wrote:
Any clue where to look for the cause?
Two questions: is there anything on the console screen? Does it just hang, or reboot?
If the latter, and there's nothing in the logs, it's possible activating openLDAP was just coincidental with the problem, and I'd start worrying about hardware problems.
- It's a VM, the console looks black. Typing something on keyboard
doesn't bring back the console to alive. It's also unresponsive to ping.
- It's just hang. Not reboot. I have to power off the VM and power on.
The ESXi host has many VM in it and only my that VM has problem.
No specific time of hang.
Only thing I can try is to deactivate auditd. Let's see if it survives more than one day without hang. _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Try to using kdump for more information about your problem. I've experience with some problem which is I can't see in /var/log/messages. So I use kdump for what truly happening in my machine.
Regards, Adit
- It's just hang. Not reboot. I have to power off the VM and power on.
The ESXi host has many VM in it and only my that VM has problem.
If you can't trace your problem to anything else, then I would look at the ESXI configuration for that VM. If there are other CentOS/Redhat 5 VM's on the ESXI server, check the vmware configuration to see that they are the same. In particular, make sure the OS is set to the correct thing for Redhat 5. VMware, in its hardware emulation, makes assumptions about clocking and possibly other kernel options and if the VM config does not agree with the running kernel, you can have problems.
Nataraj
On Thu, Nov 10, 2011 at 5:03 PM, Fajar Priyanto fajarpri@arinet.org wrote:
Hi all, Recently one my Centos 5.7 VM just crashes at least once a day randomly (hang).
In /var/log/messages there is nothing at all that there is problem (no error, no failure). The log just stops.
The only change I did before this crashes is I activated LDAP authentication, and also auditd. But I don't see any evidence relating to it.
Hi all, I want to give update.... Removing auditd didn't solve the problem. Still hang.
So, finally I create a cronjob to echo date and time into a text file every 5 minutes to find out if there is a pattern on when the hang occurs and the result is ........ It hasn't hang for more than 3 days...!
Definitely there is still problem, but somehow that cronjob prevents the hang from occurring. Like as if the VM got idle, it will hang? I don't know...
Vreme: 11/14/2011 05:09 PM, Fajar Priyanto piše:
On Thu, Nov 10, 2011 at 5:03 PM, Fajar Priyantofajarpri@arinet.org wrote:
Hi all, Recently one my Centos 5.7 VM just crashes at least once a day randomly (hang).
In /var/log/messages there is nothing at all that there is problem (no error, no failure). The log just stops.
The only change I did before this crashes is I activated LDAP authentication, and also auditd. But I don't see any evidence relating to it.
Hi all, I want to give update.... Removing auditd didn't solve the problem. Still hang.
So, finally I create a cronjob to echo date and time into a text file every 5 minutes to find out if there is a pattern on when the hang occurs and the result is ........ It hasn't hang for more than 3 days...!
Definitely there is still problem, but somehow that cronjob prevents the hang from occurring. Like as if the VM got idle, it will hang? I don't know... _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Do you have some sort of standby turned on by the guest?
Or maybe Host does not see any activity of the guest and tryies to suspend it?
-
Aditya Hilman -
Fajar Priyanto -
Ljubomir Ljubojevic -
m.roth＠5-cent.us -
Nataraj