first, i found the following workaround: http://voyager-eng.livejournal.com/tag/active%20dataset and it works!
then, sun support helped me, and it seems that i hit the following bug: Zones Cloned by Using zoneadm clone Can Cause a Snapshot Name Collision When You Activate a Boot Environment (10990) The use of the zoneadm clone command to create zones can cause a snapshot name collision, which can lead to a failure or an infinite loop when you activate a boot environment. Although this bug has been fixed in this release, you may encounter this issue while upgrading from OpenSolaris 2009.06 to the Oracle Solaris 11 Express release. Workaround: Before using image-update to update your system to the Oracle Solaris 11 Express release, run the zoneadm detach command for each zone in the OpenSolaris 2009.06 boot environment. To reattach the zones: 1. Update the image. # pkg image-update 2. Boot into the boot environment that was created by image-update. 3. Run the zoneadm attach -u command for each zone in the new boot environment. i've not tested it yet, but i'm confiant. thanks for all replies, 2010/12/22 John D Groenveld <jdg...@elvis.arl.psu.edu> > In message <4d127340.5050...@ianshome.com>, Ian Collins writes: > >What happens if you detach the zones, upgrade and then reattach with -u? > > I think he's got two problems: > his ZBEs in the detach zones aren't (un)mounted properly and > his ZBEs are clones and the snapshots are confusing the hell > out of him as much as they confused the hell out of me when > I first played with that feature. > > # zonecfg -z foo > # zfs list -r |grep foo > # mount |grep foo > > We'll probably need to see the zfs properties for the ZBEs > that refuse to attach -u. > > John > groenv...@acm.org > _______________________________________________ > zones-discuss mailing list > zones-discuss@opensolaris.org >
_______________________________________________ zones-discuss mailing list zones-discuss@opensolaris.org