[CentOS] Server hangs on CentOS 5.5

Wed Mar 9 23:25:04 UTC 2011
Dr. Ed Morbius <dredmorbius at gmail.com>

on 14:31 Wed 09 Mar, Michael Eager (eager at eagerm.com) wrote:
> Dr. Ed Morbius wrote:
> 
> >If the issue is repeated but rare system failures on one of a set of
> >similarly configured hosts, I'd RMA the box and get a replacement.  End
> >of story.
> 
> I'll repeat:  this is a house-made system.  There's no vendor to RMA to.
> It seems obvious to me:  RMA is not a diagnostic tool.

You fab your own silicon?

I saw your reference to a homebrew machine after I'd posted.  You'd
neglected to provide this information initially.

Knowing some basic stuff like:  CPU architecture, memory allocation,
disk subsystem, kernel modules, etc.,
 
> >If you'd post
> >details of the host, more logging information, netconsole panic logs,
> >etc., it might be possible to narrow down possible causes.
> 
> The problem is that there are NO DIAGNOSTICS generated when the
> system hangs.  There's no panic and nothing in the logs which
> indicates any problem.  This is what I indicated from the get go.

uname -a
/proc/cpuinfo
/proc/meminfo
lspci
lsmod
/proc/mounts
/proc/scsi/scsi
/proc/partitions
dmidecode

... would be useful for starters.

If you've built your own kernel, your config options (if you're running
stock, we can get that from the package itself).

As would wiring up netconsole as I initially suggested.


If I can clarify:  YOU are the person with the problem.  WE are the
people you're turning to for assistance.  YOU are getting pissy.  YOU
should be focusing on providing relevant information, or noting that
it's not available.

You're NOT obliged to repeat information you've already posted (e.g.:
home-brew system), but it's helpful to front-load data rather than have
us tease it out of you.
 
> >With what you've posted to date, it's not.
> 
> I could waste my time posting logs for you to tell me that they don't
> point to any problem.  I'd rather skip that step.

Krell forfend you should post relevant and useful information which
might be useful in actually diagnosing your problem (or pointing to
likely candidates and/or further tests).

-- 
Dr. Ed Morbius, Chief Scientist /            |
  Robot Wrangler / Staff Psychologist        | When you seek unlimited power
Krell Power Systems Unlimited                |                  Go to Krell!