[OmniOS-discuss] NFS Datastore vmware esxi failover

[Chris Siebenmann]

| I am working on a failover script using OmniOS as a NFS server.
[...]
| I am using zrep to failover my zfs dataset.
| http://www.bolthole.com/solaris/zrep/zrep.documentation.html
|
| Essential, it puts primary into read-only, does a zfs send/receive,
| then sets the secondary to rw.

 Transparent NFS failover is extremely demanding about the two versions
of the filesystems being absolutely identical down to extremely low
level details. It's not clear to me if zfs send and zfs recv replicate
all of these low-level details; they are certainly not documented to do
so.

(At the technical level, the two NFS servers must use identical NFS
filehandles for all filesystem objects. One problem is that part of NFS
filehandles for ZFS filesystems is an 'objset unique ID' and as far as
I can tell the ultimate source of this objset unique ID is *not* set by
zfs recv; it is set to a unique value by the kernel when the objset is
created.)

	- cks