public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* How to diagnose...
@ 2002-10-29 15:27 Robert L. Harris
  2002-10-29 15:42 ` Mikael Pettersson
  0 siblings, 1 reply; 2+ messages in thread
From: Robert L. Harris @ 2002-10-29 15:27 UTC (permalink / raw)
  To: Linux-Kernel



  Need some brainstorming from people with a clue.

Hardware:
  4xP3-550
  16Gig of Ram
  1x18Gig internal disk
  4x54Gig RAID Disk
  512Meg swap partition on sda2
  1Gig swap file in /usr/local/swapfile on /dev/sda3

The 4 54Gig disks are in a Raid5 with software raid.

I'm on a vanila 2.4.18 kernel configured for Huge Memory.

This is supposed to be a new corporate mail server but we're having some
issues.  What we have been see'ing the 3-4 times we tried is the Load 
jumps to 9+ and the box drops to a crawl when we rsync the imap folders 
from the old host to the new host.  Last night I readded the 1Gig file 
(it was at 512Meg only) and we started again.  It ran great for about 
2hrs then the box locked up.  I got to the console this morning and it 
was scrolling so fast I couldn't even read it.  It looked like it was 
reporting ACIC errors on a CPU but couldn't quite be sure.  It required 
a hard reset as it was unresponsive to c-a-d and sysreq commands.

There is nothing in the messages file and there was nothing useful on
the console.

"free" shows the rsync eats up alot of memory but never starts to swap
or if it does only swaps less than 30k.


Thoughts?


:wq!
---------------------------------------------------------------------------
Robert L. Harris                
                               
DISCLAIMER:
      These are MY OPINIONS ALONE.  I speak for no-one else.
FYI:
 perl -e 'print $i=pack(c5,(41*2),sqrt(7056),(unpack(c,H)-2),oct(115),10);'


^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: How to diagnose...
  2002-10-29 15:27 How to diagnose Robert L. Harris
@ 2002-10-29 15:42 ` Mikael Pettersson
  0 siblings, 0 replies; 2+ messages in thread
From: Mikael Pettersson @ 2002-10-29 15:42 UTC (permalink / raw)
  To: Robert L. Harris; +Cc: Linux-Kernel

Robert L. Harris writes:
 > was scrolling so fast I couldn't even read it.  It looked like it was 
 > reporting ACIC errors on a CPU but couldn't quite be sure.  It required 
 > a hard reset as it was unresponsive to c-a-d and sysreq commands.

If it says "APIC error on CPU blah" then you have a problem.
First, upgrade to something a bit more trustworthy, like 2.4.20-pre11
or RedHat's 2.4.18-17.7.x kernel. If that doesn't help, try booting
with "noapic" as parameter to the kernel.

/Mikael

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2002-10-29 15:36 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2002-10-29 15:27 How to diagnose Robert L. Harris
2002-10-29 15:42 ` Mikael Pettersson

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox