Hi, I found an issue with cl_status/hblinkstatus when it is used with multicast. The cluster behaves as expected, but cl_status reports a dead link on the host where it runs:
[r...@xen-20 ~]# checkhblinks
xen-20 br0: dead
xen-21 br0: up
xen-22 br0: up
[r...@xen-21 ~]# checkhbstatus
xen-20 br0: up
xen-21 br0: dead
xen-22 br0: up
[r...@xen-22 ~]# checkhbstatus
xen-20 br0: up
xen-21 br0: up
xen-22 br0: dead
[r...@xen-20 ~]# crm_mon -1
============
Last updated: Wed Nov 10 13:52:38 2010
Stack: Heartbeat
Current DC: xen-22 (a876c26d-00b9-437c-8e88-b19517bdf3d7) - partition with
quorum
Version: 1.0.9-74392a28b7f31d7ddc86689598bd23114f58978b
3 Nodes configured, unknown expected votes
4 Resources configured.
============
Online: [ xen-20 xen-21 xen-22 ]
stonith-xen-20 (stonith:external/ipmi): Started xen-22
stonith-xen-21 (stonith:external/ipmi): Started xen-22
stonith-xen-22 (stonith:external/ipmi): Started xen-21
Clone Set: connected
Started: [ xen-22 xen-21 xen-20 ]
This doesn't happen for bcast links, only mcast.
iptables/selinux are disabled, heartbeat 3.0.3
I have attached checkhbstatus script and ha.cf
Thank you,
Vadym
ha.cf
Description: Binary data
checkhbstatus
Description: Binary data
_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
