LVS
lvs-users
Google
 
Web LinuxVirtualServer.org

Re: Bizarre LVS oddity - one VIP handled find, another givesip_rt_bug er

To: lvs-users@xxxxxxxxxxxxxxxxxxxxxx
Subject: Re: Bizarre LVS oddity - one VIP handled find, another givesip_rt_bug errors
From: Graeme Fowler <graeme@xxxxxxxxxxx>
Date: Thu, 1 Dec 2005 12:56:25 +0000
On Thu 01 Dec 2005 12:48:28 GMT , John Line <jml4@xxxxxxxxxxxxxx> wrote:
We're not using NAT, but the current SLES9 kernel is 2.6.5-7.201, which could be enough to include old bugs for which SuSE's not back-ported fixes.

Hrm...

Here's a random idea: if you don't have the lvs debug sysctl in /proc (/proc/sys/net/ipv4/vs/debug_level) then recompile your kernel - not too familiar with SLES I'm afraid, so no pointers there :) - and ensure that IP_VS_DEBUG=y.

Reboot the director with the debug kernel, turn on debugging (start with 'echo "3" >> /proc/sys/net/ipv4/vs/debug_level' and work up or down according to how much detail you want) and follow the logs.

It should give you an idea of why your kernel is exhibiting the bug; it can however be *very* noisy (I ran a system at debug 5 for one hour last year and generated a phenomenal amount of data, hundreds of MB) so be prepared for a hunt. Hopefully you'll see why the ip_rt_bug is being hit, or at the very least *where* the transactions are failing in relation to normality. That way you might find out which additional patch is needed.

Graeme


<Prev in Thread] Current Thread [Next in Thread>