This content has been marked as final. Show 2 replies
I believe it uses zpools instead of just filesystems, because zpools unlock migration (i.e zpool export and import) which nested filesystems don't easily give you (especially if that zpool is on NFS, making zone migration very easy). And the disk names contain unique identifiers... so even if you had 2 people creating a zone called "myZone" on 2 different systems... Ops Center would give unique zpool names to each to avoid namespace collisions.
If that's not right, somebody please correct me, thanks!
After some weeks and many problems I have finally been able to create a Solaris 11 zone using Op Center 12c Release 1 (126.96.36.199.0). In doing so I have found some annoying restrictions as follows:
1. It seems that all guest zone names must end in a numeral.
2. By default guest zone filesystems get created under /guests on the global zone (when using the default local storage library).
3. I've found no way of overriding the ugly zfs file naming convention that Op Center uses for guest zones (the subject of this thread).
Has anyone found a workaround to any of these restrictions?