Re: [CentOS] centos 4.3 on esx server high load early hours carn't connect.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]



On Tue, 2006-08-22 at 10:28 +0100, Andy Dean wrote:
> Hi
> 
> posted this on the forum but then found the mailing list :-) so hope
> it's ok to post to the list aswell.
> 
> we have a centos 4.3 server running samba, winbind and squid
> authenticating by winbind to our NT domain. Once in a while in the
> early hours it has the below error in /var/log/messages this causes
> the machine to fail and we carn't connect to it. It's double dutch to 
> me, hopefully someone on here will have a suggestion though. TIA
> 
> Aug 22 04:03:15 wssproxy kernel: Unable to handle kernel NULL pointer
> dereference at virtual address 00000024

Seems to be sort of an epidemic. Check for hardware malfunction. I am on
my old reliable AMD K6-III as I write this due to a long pursued and
very painful intermittent hardware failure that begins essentially with
what you have.

I well deserved lesson about the conspiracy that exists between
hardware, software and the "Computing Gods of Torture" has been
administered to me.

A nul pointer can be due to a software error, but also caused by
hardware failure, usually a memory error I think.

Get the room hot and run memtest86 overnight.

Mine was the congruence of (I think) battery backup failure,
thunderstorms, software updates and idiocy in the wetware here. Took me
about two months to realize that it was likely hardware. I do owe a
[SOLVED] post after I cursed seamonkey updates, which occurred during my
trials.


> Aug 22 04:03:15 wssproxy kernel: printing eip:
> Aug 22 04:03:15 wssproxy kernel: c0187833

If it's *not* hardware, this eip can be used with other information by a
programmer to determine what function and instruction was in process.
All the below might be useful. But likely only if you have repeatability
and post a bug.

If it just happens occasionally and has inconsistent behavior, it'll be
difficult to chase down. And is more likely hardware after the software
has matured.

> <snip>

Another one you might see is the message about glibc detecting double
frees on software that has been running unchanged for a long time.

> Cheers
> Andy Dean
> <snip dumb "If you read this I have to kill you company blurb>

HTH
-- 
Bill

Attachment: signature.asc
Description: This is a digitally signed message part

_______________________________________________
CentOS mailing list
CentOS@xxxxxxxxxx
http://lists.centos.org/mailman/listinfo/centos

[Index of Archives]     [CentOS]     [CentOS Announce]     [CentOS Development]     [CentOS ARM Devel]     [CentOS Docs]     [CentOS Virtualization]     [Carrier Grade Linux]     [Linux Media]     [Asterisk]     [DCCP]     [Netdev]     [Xorg]     [Linux USB]
  Powered by Linux