|
Hi Guys
A while ago, I reported that I was having a problem, whereby ksoftirqd was
using 100% of the CPU (
http://lists.graemef.net/pipermail/lvs-users/2012-May/024583.html ).
Given that some reported that a later kernel, fixed their issue. I proceeded to
use Debians backport kernel (i.e. 3.2.0-0.bpo.2-686-pae) on just one of the
nodes.
But now there is a new problem. If I restart heartbeat and try to failover to
the node running the backport kernel. Heartbeat wont start the virtual ip, and
going through the logs, I see the following message.
ResourceManager[14721]: 2012/06/20_10:51:04 info: Retrying failed stop
operation [LVSSyncDaemonSwap::master]
ResourceManager[14721]: 2012/06/20_10:51:04 info: Running
/etc/ha.d/resource.d/LVSSyncDaemonSwap master stop
ResourceManager[14721]: 2012/06/20_10:51:05 ERROR: Return code 255 from
/etc/ha.d/resource.d/LVSSyncDaemonSwap
ResourceManager[14721]: 2012/06/20_10:51:05 ERROR: Resource script for
LVSSyncDaemonSwap::master probably not LSB-compliant.
ResourceManager[14721]: 2012/06/20_10:51:05 WARN: it
(LVSSyncDaemonSwap::master) MUST succeed on a stop when already stopped
ResourceManager[14721]: 2012/06/20_10:51:05 WARN: Machine reboot narrowly
avoided!
I am running Debian Squeeze therefore the ipvsadm version is:
ipvsadm 1:1.25.clean-1
I went to http://www.linuxvirtualserver.org/software/ipvs.html, and it says:
ipvsadm-1.25.tar.gz (for kernel 2.6.28-rc3 or later) - November 5, 2008
I see there is a ipvsadm-1.26-1, but I cant see anything in the changelog, for
the last entry appears to be
2004-12-08 22:55:27-08:00
Would anyone be able to help.
Kind Regards
Brent Clark
_______________________________________________
Please read the documentation before posting - it's available at:
http://www.linuxvirtualserver.org/
LinuxVirtualServer.org mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
Send requests to lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
or go to http://lists.graemef.net/mailman/listinfo/lvs-users
|
