|
On Mon, 2010-09-13 at 22:13 +0100, JL wrote:
> ...I am familiar with the scenario you are describing - where each
> machine decides the other should handle the connection - but it is not
> the situation that is occurring at the moment. This problem didn't
> happen with the exact same setup under 2.6.27.45, and...
Hrm.
Then might I recommend an Occam's Roazor approach to this? Compile a
number of kernels from the one which worked up to the one which doesn't.
Reboot the backup; test; if it doesn't happen then you know the change
cam later than the kernel you're running.
> Thanks, but I don't think we have cracked it yet. (However, for
> thoroughness, I will remove the fwmark rules on the backup, as you
> suggested, and report back.)
Not the fwmarks! For completeness I meant "clear out the IPVS rules on
the backup but leave the modules loaded".
Graeme
signature.asc
Description: This is a digitally signed message part
_______________________________________________
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
| 