[Pacemaker] Newbie question

[Mark Schenk]

Thanks for the pointer Andrew. It turns out I needed to add these 
statements:

location mysql1fencing fencing 200: mysql1
location mysql2fencing fencing 200: mysql2

So as to allow the fencing to actually run on these nodes (I am running 
with cluster-option symmetric-cluster="false", but I forgot to mention that)

Mark

PS. Dominik thanks for trying, my info probably should have been clearer...

Andrew Beekhof wrote:
> Try:
>     crm_verify -V
>
> It should give you a warning about stonith.
>
> On Thu, May 14, 2009 at 2:36 PM, Mark Schenk <M.M.A.Schenk at tudelft.nl> wrote:
>   
>> Hello All,
>>
>>   I'm new to pacemaker so please forgive me if this is in a faq somewhere, I
>> haven't been able to find it!
>> I am trying to set up a failover config for mysql using the following setup:
>>
>> primitive mysqlvip ocf:heartbeat:IPaddr params ip="xx.xx.xx.xx"
>> primitive mysqlfs ocf:heartbeat:Filesystem params device="-L mysql"
>> directory="/var/lib/mysql" fstype="ext3"
>> primitive mysqlserver ocf:heartbeat:mysql params
>> binary="/usr/bin/mysqld_safe" op monitor interval="60s" timeout="30s"
>> primitive st-null stonith:null params hostlist="mysql1 mysql2"
>> group mysqlservice mysqlvip mysqlfs mysqlserver
>> clone fencing st-null
>> location mysql1loc mysqlservice 200: mysql1
>> location mysql2loc mysqlservice 150: mysql2
>> colocation mysql_ip_with_fs inf: mysqlvip mysqlfs
>> colocation mysql_server_with_fs inf: mysqlserver mysqlfs
>> order mysql_fsbeforeserver inf: mysqlfs mysqlserver
>>
>> And it looks to be working great. When I start the configuration three
>> resources are started on node mysql1. When I switch off node1 by shutting
>> down the heartbeat service I can see all my resources moving over to node
>> mysql2. However when I pull the plug on node mysql1 by switching it off, my
>> services are not moving over. Instead my services are listed as having been
>> started on node mysql1 (which is down!). This is the output of crm_mon -r:
>>
>>
>> ============
>> Last updated: Thu May 14 14:32:46 2009
>> Current DC: nfs2 (e588fc86-92d8-4658-a394-3e031171bcd0) - partition with
>> quorum
>> Version: 1.0.3-b133b3f19797c00f9189f4b66b513963f9d25db9
>> 6 Nodes configured, unknown expected votes
>> 2 Resources configured.
>> ============
>>
>> Node mysql1 (ea3da999-829d-4042-a5d5-9ee0ad27f989): UNCLEAN (offline)
>> Online: [ apa1 apa2 mysql2 nfs1 nfs2 ]
>>
>> Full list of resources:
>>
>> Resource Group: mysqlservice
>>     mysqlvip    (ocf::heartbeat:IPaddr):        Started mysql1
>>     mysqlfs     (ocf::heartbeat:Filesystem):    Started mysql1
>>     mysqlserver (ocf::heartbeat:mysql):   Started mysql1
>> Clone Set: fencing
>>         Stopped: [ st-null:0 st-null:1 st-null:2 st-null:3 st-null:4
>> st-null:5 ]
>>
>> I must be missing something but I haven't yet found it.
>>
>> Any pointers would be appreciated.
>>
>> Thanks,
>>   Mark
>>
>> --
>> _____________________________________________________________________
>> Mark Schenk
>> Project Manager
>> Shared Service Centre ICT - 3xO
>> Technische Universiteit Delft
>> M.M.A.Schenk at tudelft.nl
>> Room 3B-41
>> Postbus 354
>> 2600 AJ Delft
>> Landbergstraat 15
>> 2628 CE DELFT             ___________________________________________
>> +31 (0)15 27 85448        | Reality is that which, when you
>> www.tudelft.nl            | stop believing in it, doesn't go away.
>> _____________________________________________________________________
>>
>> _______________________________________________
>> 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
>   

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://oss.clusterlabs.org/pipermail/pacemaker/attachments/20090514/6bfe0f1a/attachment.htm>