Thanks! I still feel like I'm missing two vital pieces of the puzzle. First, for better or worse, the sparse zones were created with separate /usr dirs, using the command:
zonecfg:zone_1> remove inherit-pkg-dir dir=/usr Also, I have four zones on one machine (and ten on another!) The four zones are sharing one physical partition, named /space. I don't have enough free partitions to make one for each zone, and Solaris 10 u3 doesn't allow for ZFS roots so I can't use my ZFS SAN partition (another reason to upgrade!) So do I understand correctly that if all my zones live on /space, If I do: lucreate -n newbe -m /space:/dev/dsk/whatver:/ufs liveupgrade will duplicate /space in the new BE and use it for all of the zones that live on /space? And just to complicate matters, on the test server, /space is a heck of a lot bigger than any other free partition. Would this work if I lived dangerously, allow my alternate boot environment to mount /space, and let it update the zones? Recognizing that if the ugrade fails, my zones are toast and I have no rollback (on this particular server, rebuilding the zones would be relatively easy) (I'm obviously going to learn a lot that I can use to redesign my next generation of servers, but meanwhile I'm trying to drag this group into this decade. And the irony is that, when this is all done, I'll have a server I can use as a *proper* test server. And double irony, I'll be at an OS release that'll let me use ZFS root file systems, but I'm not there yet...) thanks for your patience, this is all sort of an emergency because I've got a production server crashing and Sun is insisting that it needs an upgrade or bust - my original plan to do this over a 2-month period and spend lots of time with a test server, is toast. _______________________________________________ zones-discuss mailing list zones-discuss@opensolaris.org
