[ClusterLabs] resource grouping

[Kristoffer Grönlund]

Dennis Jacobfeuerborn <dennisml at conversis.de> writes:

> Hi,
> is there a way to group resources without also adding implicit ordering?
>
> My use-case is that I want to set up a redundant NFS system that uses
> drbd to mirror a few volumes. These volumes are then mounted and exported.
> The problem is that if I have e.g. 10 Filesystem resources I currently
> have to both a colocation and an order constraint for *each* filesystem
> which is quite cumbersome and as a result also prone to errors.
>
> What I'd like to be able to do is to put all filesystem resources in a
> group and then only create one colocation and order contraint each
> between the drbd clone resource and this filesystem resource group.
> Since right now grouping implies ordering that is somewhat inefficient
> because the filesystems cannot be mounted in parallel when grouped.
>
> Is there a way to accomplish what I'm trying to do in pacemaker/pcs?

Yes, there is something called tags. It is like a group, but doesn't
imply any constraints. You can then apply constraints to the tag. See
http://crmsh.github.io/man/#cmdhelp_configure_tag

This is a new feature in pacemaker 1.1.12+.

>
> Regards,
>   Dennis
>
> _______________________________________________
> Users mailing list: Users at clusterlabs.org
> http://clusterlabs.org/mailman/listinfo/users
>
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs: http://bugs.clusterlabs.org
>

-- 
// Kristoffer Grönlund
// kgronlund at suse.com