[Pacemaker] Shared Storage

[Richard de Vries]

Hello Ryan,

glusterfs is not the same as drdb. We use it to replicate data from
one node to the other and use corosync/pacemaker to do guaranty that
one node is active.

I assume that you can never directly touch the R2, because there is a
filesystem on R1 that is replicated to R2?

Regards,
Richard

> Thanks for the reply, Richard
>
>I went to the site (gluster.org). It seems it runs the same way as DRBD. In EMC's word, the primary storage is called R1 and replicate copy of the storage is called R2.
>DRBD is node A of the cluster accesses and controls R1. Node B of the cluster accesses R2. That is what my understanding. Is gluster the same as DRBD?
>HP's HPUX MC/ServiceGuard is both node A and B of the cluster access R1. Normally node A controls R1. When it fails over, node B takes control R1 (active/standby). Is there similar software exists? >Redhat's cluster is similar but node A and B have control of R1 (active-active).
>
>Ryan