On Tue, Jul 1, 2008 at 6:43 PM, <Madhan.Balasubramanian at sun.com> wrote: > Hi Moinak, > > Since I had created the snapshot at the zpool level, I had expected it to > work with -r option when executing test at complete. > > Executing a rollback @ /test/testfs/1 level worked. > > So wondering what is the use of -r option as testfs and testfs/1 descend > from the test.
Just checked the manpage. '-r' for taking a snapshot will recursively traverse all child datasets and take individual snapshots for each. '-r' when doing a rollback is different. While rolling back to a snapshot it will rollback and destroy all snapshots newer than the one specified (if they exist) for the given dataset. It will not recurse into child datasets. Regards, Moinak. > > Best Regards, > Madhan Kumar > > Moinak Ghosh wrote: > > On Tue, Jul 1, 2008 at 6:24 PM, <Madhan.Balasubramanian at sun.com> wrote: > > > Moinak, > > That's where confusion sets in .. > > In this case I have only one dataset which is complete and for all the child > FS of the zpool test. > > > That is not the case. If you see the the zfs list carefully you have 3 > independent snapshots, one for the zfs dataset for the pool and one > each for the 2 additional child datasets. The pool itself has a default > zfs dataset that is mounted under /test. > > Regards, > Moinak. > > > _______________________________________________ > ug-bosug mailing list > List-Unsubscribe: mailto:ug-bosug-unsubscribe at opensolaris.org > List-Owner: mailto:ug-bosug-owner at opensolaris.org > List-Archives: http://www.opensolaris.org/jive/forum.jspa?forumID=54 >
