[Linux-HA] Issue with drbd filesystem unmanaged in CRM

Lars Marowsky-Bree lmb at suse.de
Fri Sep 8 07:20:40 MDT 2006


On 2006-09-14T22:56:58, Lewis Shobbrook <mylists at blue-matrix.org> wrote:

> I've migrated a working system to crm in hbeat 2.07-1 and have a problem with 
> a drbd resource in one of the groups.  
> Basically the resource lists as unmanaged and the ocf filesystem resource 
> beneath it in the group lists as not started. 

The drbd/class=ocf resource agent is a master-slave resource and needs
to be configured as such. As m/s resources are currently not very useful
due to limitations in the Policy Engine, you're better off using the
drbd init script + drbddisk method.

For this, setup drbd as you'd have with heartbeat 1.x, and use the
drbddisk/datadisk provided by the drbd package as a primitive resource.

"drbd" from the OCF scripts versus the drbddisk script take entirely
different parameters. You can't simply change the name & class and
expect this to work!

> Also found a non responsive heartbeat to a restart call and was forced to 
> killall => corrupt cib.xml 8=(

The CIB should not get corrupted. What is the bugzilla id here?

> Think I might have to leave the GUI & crm_resource manager and head back at 
> this rate....

Various companies sell consulting if you don't want to figure it out
yourself ;-)



Sincerely,
    Lars Marowsky-Brée

-- 
High Availability & Clustering
SUSE Labs, Research and Development
SUSE LINUX Products GmbH - A Novell Business	 -- Charles Darwin
"Ignorance more frequently begets confidence than does knowledge"



More information about the Linux-HA mailing list