On 27/03/2019 21:27, Valentin Vidic wrote:
On Wed, Mar 27, 2019 at 08:13:07AM +0100, Ulrich Windl wrote:
Seems to be traditional. Maybe it's just to create some namespace.
Yes, I think GFS2 and OCFS2 store cluster name in FS metadata.
If the cluster name changes they might not mount anymore:
https://access.redhat.com/solutions/18430
If the Filesystem resource agent was smarter it could figure out the
cluster name when it's enabled and use the -o
locktable=clustername:fsname mount option, but that has its own set of
drawbacks of course.
Andy
_______________________________________________
Manage your subscription:
https://lists.clusterlabs.org/mailman/listinfo/users
ClusterLabs home: https://www.clusterlabs.org/
