Not 100% sure but if ib0 and ib1 are suppose to be redundant, I'd just
create a bonded interface of them and call it ibx, let the failover of
ib pipeline be handled at the driver level, not corosync. /kiss/
principal. Maybe I dont have a broad enough picture but thought I'd
offer my humble opinion.
On 12/16/2013 7:41 AM, Frank Kautz wrote:
Hello,
I have a configuration with 3 network interfaces (eth0, ib0, ib1). The
interface eth0 is configured for ring 0. ib0 and ib1 are configured for
ring 1 because both interfaces are in the same IP subnet. It looks like
corosync uses every time the first interface ib0. What happened when the
first interface (ib0) is down? Do corosync use the second interface
(ib1)? Or does corosync ignore the second interface in the same ring?
kind regards
Frank
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
