[lvs-users] LVS RR becomes unbalanced after time
victor at sartori.eti.br
Mon Jul 4 18:02:44 BST 2011
looking on my logs, i found this:
keepalived: segfault at 0 ip 0000000000407f6b sp 00007fffd2e5f0d0
error 6 in keepalived[400000+29000]
Did you have the same segfault too?
2011/7/4 Luc van Donkersgoed <luc at codingdutchmen.com>
> Well, just ploughing on, I came to the conclusion that because there are no
> ActiveConn and InactiveConn, the total number of connections (as displayed
> in ipvsadm -ln --stats) does not get updated. This means there is no trigger
> for the loadbalancer to send requests to the next server, which results in
> all the request being sent to the same realserver.
> The root of the problem therefore seems to be that the ActiveConn and
> InactiveConn don't get updated, even while new clients are connecting.
> Please read the documentation before posting - it's available at:
> LinuxVirtualServer.org mailing list - lvs-users at LinuxVirtualServer.org
> Send requests to lvs-users-request at LinuxVirtualServer.org
> or go to http://lists.graemef.net/mailman/listinfo/lvs-users
More information about the lvs-users