[Pacemaker] Errors in configuring cluster project with DRBD and OCFS2, Help !

Thu Jun 20 18:02:45 CEST 2013

----- Original Message -----
> From: "Jonathan-Marc Lapointe" <jmlapointe at ccdmd.qc.ca>
> To: "Jake Smith" <jsmith at argotec.com>, "The Pacemaker cluster resource manager" <pacemaker at oss.clusterlabs.org>
> Sent: Thursday, June 20, 2013 8:27:07 AM
> Subject: Re: [Pacemaker] Errors in configuring cluster project with DRBD and OCFS2, Help !
> 
> Thanx for your response.
> 
> I tried to do What you said, I inverted the resource in the
> colocation statement but it did not fix the problem.
> 
> So I deleted the resources one after the other until I had only 2
> resources.
> 
> Then It refused to remove them saying that they where started but I
> could not stop them and the "crm resource cleanup" command did not
> allow me to remove them.
> 

Did you try to cleanup the ms or primitive resource?  Should cleanup on the ms_drbd_ocfs2.

If that doesn't work (I'm assuming this is a test environment?) shutdown both nodes then restart - clears cluster status and starts over.

Do you have stonith configured?

> Here is my config now:
> 
> -------------------------------------------------------------------------------------
> 
> primitive p_drbd_ocfs2 ocf:linbit:drbd \
> params drbd_resource="r0"
> ms ms_drbd_ocfs2 p_drbd_ocfs2 \
> meta master-max="2" clone-max="2" target-role="Started"
> is-managed="true"
> 
> -------------------------------------------------------------------------------------
> 
> But I still have the following error message:
> 
> -------------------------------------------------------------------------------------
> 
> Master/Slave Set: ms_drbd_ocfs2 [p_drbd_ocfs2]
> p_drbd_ocfs2:0 (ocf::linbit:drbd): Slave web1dev (unmanaged) FAILED
> p_drbd_ocfs2:1 (ocf::linbit:drbd): Slave web2dev (unmanaged) FAILED
> 
> Failed actions:
> p_drbd_ocfs2:1_stop_0 (node=web2dev, call=4, rc=6, status=complete):
> not configured
> p_drbd_ocfs2:0_stop_0 (node=web1dev, call=4, rc=6, status=complete):
> not configured
> 

These are usually from attempting to stop a resource and failing to succeed at stopping -> ends up failed and unmanaged.

Logs should tell you why they failed to stop.

HTH

Jake

> -------------------------------------------------------------------------------------
> 
> Could you or anyone explain me the errors meaning?
> 
> Thanx !
> 
> 
> 
> Le 2013-06-19 16:16, Jake Smith a écrit :
> 
> 
> ----- Original Message -----
> 
> From: "Jonathan-Marc Lapointe" <jmlapointe at ccdmd.qc.ca> To:
> Pacemaker at oss.clusterlabs.org Sent: Wednesday, June 19, 2013 8:21:03
> AM
> Subject: [Pacemaker] Errors in configuring cluster project with DRBD
> and OCFS2, Help !
> 
> Hi,
> 
> I have a project I'm working on and I'm stuck right now.
> 
> I'm building a cluster of 2 load balanced web server and I'm using
> DRBD, OCFS2 and Pacemaker/Corosync.
> 
> My Problem is with configuring Pacemaker, the rest is working fine.
> 
> Here is the config I used to configure my resources:
> 
> --------------------------------------------------------------------------------
> primitive p_drbd_ocfs2 ocf:linbit:drbd \
> params drbd_resource="ocfs2"
> ms ms_drbd_ocfs2 p_drbd_ocfs2 \
> meta master-max=2 clone-max=2
> primitive p_controld ocf:pacemaker:controld
> primitive p_o2cb ocf:pacemaker:o2cb
> group g_ocfs2mgmt p_controld p_o2cb
> clone cl_ocfs2mgmt g_ocfs2mgmt \
> meta interleave=true
> primitive p_fs_ocfs2 ocf:heartbeat:Filesystem \
> params device="/dev/drbd0" directory="/mnt/data" fstype="ocfs2"
> options="rw,noatime"
> clone cl_fs_ocfs2 p_fs_ocfs2
> order o_ocfs2 inf: ms_drbd_ocfs2:promote cl_ocfs2mgmt:start
> cl_fs_ocfs2:start
> colocation c_ocfs2 inf: cl_fs_ocfs2 cl_ocfs2mgmt ms_drbd_ocfs2:Master
> Pretty sure your problem is this colocation statement.  Try running
> crm configure show xml c_ocfs2 and see if the order of cl_fs and
> cl_ocf2mgmt is reversed from what it should be. When you have more
> than two resources in a colocation statement they become a set and
> the order of a set is reversed.  This means the colocation is saying
> that controld must run with o2cb must run with filesystem must run
> with drbd:Master.
> 
> You could just add p_fs_ocfs2 to the end of the g_ocfs2mgmt group
> instead of being separate (my preference) or reverse the first two
> resources in the colocation (knowing they are parsed the other way
> around) or you could write the colocations in two statements with
> only two resources each.
> 
> HTH
> 
> Jake
> 
> --------------------------------------------------------------------------------
> 
> But when I commit it, I get the following errors in crm_mon that I
> don't understand:
> 
> --------------------------------------------------------------------------------
> ============
> Last updated: Tue Jun 18 16:36:02 2013
> Last change: Tue Jun 18 16:36:02 2013 via crm_shadow on web1dev
> Stack: openais
> Current DC: web1dev - partition with quorum
> Version: 1.1.6-9971ebba4494012a93c03b40a2c58ec0eb60f50c
> 2 Nodes configured, 2 expected votes
> 8 Resources configured.
> ============
> 
> Online: [ web2dev web1dev ]
> 
> Master/Slave Set: ms_drbd_ocfs2 [p_drbd_ocfs2]
> p_drbd_ocfs2:0 (ocf::linbit:drbd): Slave web1dev (unmanaged) FAILED
> p_drbd_ocfs2:1 (ocf::linbit:drbd): Slave web2dev (unmanaged) FAILED
> 
> Failed actions:
> p_drbd_ocfs2:0_stop_0 (node=web1dev, call=10, rc=6, status=complete):
> not configured
> p_controld:0_monitor_0 (node=web1dev, call=6, rc=5, status=complete):
> not installed
> p_drbd_ocfs2:1_stop_0 (node=web2dev, call=10, rc=6, status=complete):
> not configured
> p_controld:0_monitor_0 (node=web2dev, call=6, rc=5, status=complete):
> not installed
> --------------------------------------------------------------------------------
> 
> Any help appreciated !
> 
> Thanx !
> 
> CCDMD 20 ans !
> 
> Jonathan-Marc Lapointe
> Technicien en informatique,
> Administrateur système,
> Administrateur réseau
> 
> Centre collégial de développement de matériel didactique
> 6220, rue Sherbrooke Est, bureau 404
> Montréal (Québec) H1N 1C1
> 514) 864-8443 www.ccdmd.qc.ca jmlapointe at ccdmd.qc.ca Skype: jmlccdmd
> 
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> 
> Project Home: http://www.clusterlabs.org Getting started:
> http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs:
> http://bugs.clusterlabs.org
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home:
> http://www.clusterlabs.org Getting started:
> http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs:
> http://bugs.clusterlabs.org
> 
> 
> CCDMD 20 ans !
> 
> Jonathan-Marc Lapointe
> Technicien en informatique,
> Administrateur système,
> Administrateur réseau
> 
> Centre collégial de développement de matériel didactique
> 6220, rue Sherbrooke Est, bureau 404
> Montréal (Québec) H1N 1C1
> 514) 864-8443
> www.ccdmd.qc.ca
> jmlapointe at ccdmd.qc.ca
> Skype: jmlccdmd
> 