[Pacemaker] couple of comments/questions on DRBD HowTo 1.0

[Dominik Klein]

Juha Heinanen wrote:
> Lars Ellenberg writes:
> 
>  > If that "Lars?" meant me, yes, please,
>  > go ahead an delete outdated examples. 
>  > Replace with a reference to the drbd users guide
>  > http://www.drbd.org/docs/about/ or
>  > http://www.drbd.org/docs/install/
> 
> how about the webserver example in DRBD HowTo 1.0?  i suspect that it
> does not work.

Works for me. Any other complaints?

> if i try it without 'meta target-role=stopped' in fs0 resource,
> pacemaker tried to mount the filesystem before drbd0 resource is running
> resulting in error
> 
> Mar 28 17:16:56 lenny1 lrmd: [13953]: info: rsc:fs0: start
> Filesystem[15261]:	2009/03/28_17:16:57 INFO: Running start for /dev/drbd0 on /var/lib/mysql
> Mar 28 17:16:57 lenny1 lrmd: [13953]: info: RA output: (fs0:start:stderr) /dev/drbd0: Wrong medium type

Are you sure you had the colocation and order constraints by then? If
so, please provide a hb_report covering that time so the bug can be fixed.

> (i use /var/lib/mysql instead of /usr/local/apache/htdocs).
> 
> if i then add 'meta target-role=stopped' to fs0 and give command
> 
> crm resource start ms-drbd0
> 
> mysql-group is not started automatically.  in addition i need to give
> command
> 
> crm resource start fs0
> 
> why doesn't starting of ms-drbd0 promote the whole group? it would be
> nice if the apache example would work out of the box without any
> tweaking.

Each resource has its individual target-role. It defaults to "started".
If you manually set it to "stopped", then you'll have to change it again
in order to run the resource. That's just how it works.

> my corresponding example crm config is below.  
> 
> -- juha
> 
> ------------------------------------------------------------------------
> 
> configure
> 
> primitive drbd0 ocf:heartbeat:drbd \ 
>  params drbd_resource=drbd0 \
>  op monitor role=Master interval=59s timeout=30s \
>  op monitor role=Slave interval=60s timeout=30s
> 
> ms ms-drbd0 drbd0 \
>  meta clone-max=2 notify=true globally-unique=false target-role=stopped
> 
> primitive fs0 ocf:heartbeat:Filesystem params fstype=ext3 directory=/var/lib/mysql device=/dev/drbd0 meta target-role=stopped
> 
> primitive virtual-ip ocf:heartbeat:IPaddr2 params ip=192.98.101.10 broadcast=192.98.101.255 nic=eth0 cidr_netmask=24 op monitor interval=21s timeout=5s
> 
> primitive mysql-server lsb:mysql op monitor interval=20s start-delay=30s
> 
> group mysql-server-group fs0 virtual-ip mysql-server
> 
> order ms-drbd0-before-mysql-server-group mandatory: ms-drbd0:promote mysql-server-group:start
> 
> colocation mysql-server-group-on-ms-drbd0 inf: mysql-server-group ms-drbd0:Master
> location ms-drbd0-master-on-lenny1 ms-drbd0 rule role=master 100: #uname eq lenny1
> 
> property default-resource-stickiness=1
> property failure-timeout=10s
> property cluster-recheck-interval=1m

btw: How should a failure timeout after 10s if the cluster only rechecks
every 60s?

Regards
Dominik

> commit
> 
> end
> 
> resource start ms-drbd0
> resource start fs0
> 
> quit