Hi, to enable NFS state transfer between hosts in a failover SAN, it is necessary to have the NFS state data on the exact same blocks on both storage nodes: http://serverfault.com/a/274995 <http://serverfault.com/questions/274990/fault-tolerant-nfs/274995#274995> http://sysconfig.org.uk/2011/07/glusterfs-a-workhorse-that-needs-to-be-tamed/ "You MUST have a shared (precisely the same block for block) filesystem or NFS failover won't work, and you'll get stale NFS file handles." http://lists.linux-ha.org/pipermail/linux-ha/2004-September/012005.html Simply linking /usr/lib/nfs to the glusterfs volume doesn't work. My questions: Can I somehow enable block-for-block replication with GlusterFS? What is this: http://gluster.org/community/documentation/index.php/Gluster_3.2:_Provisioning_Storage_on_Gluster_Virtual_Storage_Appliance Thank you. -- Best Regards Runar Ingebrigtsen -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://supercolony.gluster.org/pipermail/gluster-users/attachments/20121023/bf804afd/attachment.html>
