Thomas Baumann wrote:
Hello list,
I setup a 4 node cluster with drbd and mysql on node2 and node3.
But in my log I can see the monitor process for drbd will try on
node0 and node1. How can I avoid this ?
Attached you find my .dot, and my cib.xml (stripped in case
of client data).
Where is my fault ? What is missing (or even too much) ?
On node0, node1 there is no drbd (and must not!).
Thanks for your reply in advance.
Thomas.
============
Last updated: Sat Dec 6 22:45:46 2008
Current DC: node0 (e1374bf0-51b7-4535-9aba-4299898e9522)
4 Nodes configured.
3 Resources configured.
============
Node: node3 (d29dde8b-1f7a-469a-9733-a4d71a6b449e): online
Node: node2 (41969885-02ed-4fb2-913a-a79b26edfd88): online
Node: node1 (e1374bf0-51b7-4535-9aba-4299898e9522): online
Node: node0 (b4613c4b-e464-4487-bea6-0318d575352f): online
Master/Slave Set: MS_DRBD0
R_DRBD0:0 (ocf::heartbeat:drbd): Master node2
R_DRBD0:1 (ocf::heartbeat:drbd): Started node3
Resource Group: G_MySQL_DRBD0
R_DRDB0_FS (ocf::heartbeat:Filesystem): Started node2
R_DRBD0_MySQLd (lsb:mysqld): Started node2
R_VIP:213 (ocf::heartbeat:IPaddr.tiri.li): Started node2
==
Logfile:
Dec 6 22:37:06 node0 crmd: [19461]: info: process_lrm_event: LRM
operation R_VIP:211_monitor_0 (call=5, rc=7) complete
Dec 6 22:37:06 node0 crmd: [19461]: info: process_lrm_event: LRM
operation R_VIP:210_monitor_0 (call=4, rc=7) complete
Dec 6 22:37:06 node0 crmd: [19461]: info: process_lrm_event: LRM
operation R_VIP:209_monitor_0 (call=3, rc=7) complete
Dec 6 22:37:06 node0 crmd: [19461]: info: process_lrm_event: LRM
operation R_ldirector_monitor_0 (call=2, rc=7) complete
Dec 6 22:37:07 node0 lrmd: [19458]: info: rsc:R_VIP:212: monitor
Dec 6 22:37:07 node0 lrmd: [19458]: info: rsc:R_DRBD0:0: monitor
Dec 6 22:37:07 node0 lrmd: [19458]: info: rsc:R_DRDB0_FS: monitor
Dec 6 22:37:07 node0 lrmd: [19458]: info: rsc:R_DRBD0_MySQLd: monitor
Dec 6 22:37:07 node0 drbd[19780]: [19793]: WARNING: drbd0 monitor:
drbd module not loaded
Dec 6 22:37:07 node0 crmd: [19461]: info: process_lrm_event: LRM
operation R_DRBD0:0_monitor_0 (call=7, rc=7) complete
Dec 6 22:37:07 node0 crmd: [19461]: info: process_lrm_event: LRM
operation R_DRDB0_FS_monitor_0 (call=8, rc=7) complete
Dec 6 22:37:07 node0 crmd: [19461]: info: process_lrm_event: LRM
operation R_VIP:212_monitor_0 (call=6, rc=7) complete
Dec 6 22:37:07 node0 crmd: [19461]: info: process_lrm_event: LRM
operation R_DRBD0_MySQLd_monitor_0 (call=9, rc=7) complete
Dec 6 22:37:08 node0 lrmd: [19458]: info: rsc:R_VIP:179: monitor
Dec 6 22:37:08 node0 crmd: [19461]: info: process_lrm_event: LRM
operation R_VIP:179_monitor_0 (call=10, rc=7) complete
What you're seeing here is not a problem. It's the cluster testing
whether a resource is running anywhere in the cluster before starting it
elsewhere to make sure it runs okay.
The monitor operations return 7 (not running), which is expected and fine.
That's a one-time-thing before the first start of your resource, not a
recurring operation like the monitor operations on the nodes the
resource is actually running on.
Regards
Dominik
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
