[Pacemaker] Problems with colocation/order with drbdtree-node-setup

Schmidt, Torsten torsten.schmidt at tecdoc.net
Fri Oct 16 11:44:44 UTC 2009


i think there is a problem with the monitor operation for the stacked drbd resouce:

after erase + commit of my crm-config the CIP and mysqldb (lower) resource started on node 2

so i looked at drbd-status on node 2 and there was nothing to be seen of the stacked device:

/proc/drbd
--------------------------------------------------------------------
 0: cs:Connected ro:Primary/Secondary ds:UpToDate/UpToDate C r----
    ns:0 nr:30 dw:30 dr:0 al:0 bm:1 lo:0 pe:0 ua:0 ap:0 ep:1 wo:b oos:0
--------------------------------------------------------------------


ptest -L -VVVVV -s -a:
--------------------------------------------------------------------
info: res.drbd.mysqldb:0_monitor_0 on mysqlHA2 returned 0 (ok) instead of the expected value: 7 (not running)
notice: Operation res.drbd.mysqldb:0_monitor_0 found resource res.drbd.mysqldb:0 active on mysqlHA2
info: res.drbd.mysqlstack:0_monitor_0 on mysqlHA2 returned 6 (not configured) instead of the expected value: 7 (not running)
ERROR: Hard error - res.drbd.mysqlstack:0_monitor_0 failed with rc=6: Preventing ms.drbd.mysqlstack from re-starting anywhere in the cluster
info: res.drbd.mysqlstack:0_monitor_0 on mysqlHA1 returned 6 (not configured) instead of the expected value: 7 (not running)
ERROR: Hard error - res.drbd.mysqlstack:0_monitor_0 failed with rc=6: Preventing ms.drbd.mysqlstack from re-starting anywhere in the cluster 
WARN: native_color: Resource res.drbd.mysqlstack:0 cannot run anywhere
--------------------------------------------------------------------

output of crm_mon see below.

im i correct that i only have to load the drbd kernel-module on node 1 + 2; pacemaker should do the rest (configuration, promote/demote...) right?


Mit freundlichen Grüßen / with kind regards
Torsten Schmidt

-----Ursprüngliche Nachricht-----
Von: Schmidt, Torsten [mailto:torsten.schmidt at tecdoc.net] 
Gesendet: Freitag, 16. Oktober 2009 12:54
An: pacemaker at oss.clusterlabs.org
Betreff: Re: [Pacemaker] Problems with colocation/order with drbdtree-node-setup

thank you very much Johan,

i commited my configs.

the 'simple' ptest now is:
---------------------------------------------------------------------------------------------------------
ERROR: unpack_rsc_op: Hard error - res.drbd.mysqlstack:0_monitor_0 failed with rc=6: Preventing ms.drbd.mysqlstack from re-starting anywhere in the cluster
ERROR: unpack_rsc_op: Hard error - res.drbd.mysqlstack:0_monitor_0 failed with rc=6: Preventing ms.drbd.mysqlstack from re-starting anywhere in the cluster
WARN: native_color: Resource res.drbd.mysqlstack:0 cannot run anywhere
---------------------------------------------------------------------------------------------------------

looking on crm_mon:
---------------------------------------------------------------------------------------------------------
============
Last updated: Fri Oct 16 12:42:59 2009
Stack: openais
Current DC: mysqlHA1.cologne.tecdoc.local - partition with quorum
Version: 1.0.5-462f1569a43740667daf7b0f6b521742e9eb8fa7
2 Nodes configured, 2 expected votes
3 Resources configured.
============

Online: [ mysqlHA2.cologne.tecdoc.local mysqlHA1.cologne.tecdoc.local ]

Master/Slave Set: ms.drbd.mysqldb
        Masters: [ mysqlHA2.cologne.tecdoc.local ]
        Slaves: [ mysqlHA1.cologne.tecdoc.local ]
res.ip.mysql    (ocf::heartbeat:IPaddr2):       Started mysqlHA2.cologne.tecdoc.local

Failed actions:
    res.drbd.mysqlstack:0_monitor_0 (node=(null), call=4, rc=6, status=complete): not configured
    res.drbd.mysqlstack:0_monitor_0 (node=(null), call=4, rc=6, status=complete): not configured
---------------------------------------------------------------------------------------------------------

next i work through the output of ptest -L -s -VVVVV


Mit freundlichen Grüßen / with kind regards

Torsten Schmidt
System Manager

-----Ursprüngliche Nachricht-----
Von: Johan Verrept [mailto:Johan.Verrept at able.be]
Gesendet: Freitag, 16. Oktober 2009 12:38
An: pacemaker at oss.clusterlabs.org
Betreff: Re: [Pacemaker] Problems with colocation/order with drbd tree-node-setup

On Fri, 2009-10-16 at 12:32 +0200, Johan Verrept wrote:

> Other than that, I have no more ideas

perhaps ptest -L -s -VVVVV helps...


_______________________________________________
Pacemaker mailing list
Pacemaker at oss.clusterlabs.org
http://oss.clusterlabs.org/mailman/listinfo/pacemaker

_______________________________________________
Pacemaker mailing list
Pacemaker at oss.clusterlabs.org
http://oss.clusterlabs.org/mailman/listinfo/pacemaker




More information about the Pacemaker mailing list