RHEL 5.3 hangs, opportunistic locking problem?

[wobbe]

Yesterday one of our servers stoped responding.
I could still ping the server but vnc, ssh, telnet, httpd etc did not respond anymore. The console was also not working anymores so I had to restart the server.
Looking at the messages file (see attachment) we now think it might be an issue with opportunistic locking and we are considering disabling this option.

One more thing at around the same time the CIFS error message is also showing up on other servers. These servers did not stop responding.

[Ivan Ferreira]

I think that you should first do a memory test for your server. As file system cache and physical memory are related, this could be a problem with your physical memory.

For locking discussion, check this link:

http://fixunix.com/samba/342410-samba-locking-gfs.html

[wobbe]

Hi Ivan,

Why do you think it could be related to memory errors? According to the HP management agents memory is just fine. But I'm not sure how reliable these HP monitoring tests are when it comes to memory testing is always tricky.

FYI. From Redhat I got the following response:

Run following command and check whether you encounter another unresponsiveness event:

# sysctl -w kernel.softlockup_thresh=30

To make this parameter persistent across reboots by adding following line in /etc/sysctl.conf file:

# kernel.softlockup_thresh=30

Soft lockups are situations in which the kernel's scheduler subsystem has not been given a chance to perform its job for more than 10 seconds; they can be caused by defects in the kernel, by hardware issues or by extremely high workloads.