So this will actually work?
On Sat, Mar 22, 2008 at 5:58 PM, Vahid Moghaddasi <[EMAIL PROTECTED]> wrote: > Hi All, > > This is a step-by-step procedure to upgrade Solaris 10 06/06 (u2) to Solaris > 10 08/07 (u4). > Before any attempt to upgrade you will need to install 'at least' the > Recommended patches dated March 2008. > - The Kernel level of the u2 system must be 120011-14. > - The zones will have to be moved to ufs (done by LU). > - You may need to halt the zone to apply the above patches, expect downtime. > - In this document we use 'current' for the current BE and 'stage' for the > new BE. > - We assume that the zones are in /zones directory on a zfs file system. > > 1- Create a ufs file system which can hold critical file systems plus the > zones in this case c1t0d0s5. > 2- Create the new BE (stage): > # lucreate -c current -m /:/dev/dsk/c1t0d0s5:ufs -n stage > NOTE: if '-c' is omitted, the default BE name will be the device name. > 3- Upgrade the new BE (stage): > # luupgrade -u -n stage -s /net/depot/sol10sparc > 4- Activate the new BE to test: > # luactivate stage > # init 6 > NOTE: > The zfs file system in zonepath might not mount and place the > filesystem/local service in maintenance mode and complain about the zonepath > directory not being empty. In that case, you will have to move the parent of > the zonepath to another location, e.g. mv /zones /zones.u4 and run zfs mount > -a and see your original /zones file system. There will be other directories > related to zone upgrade in /zones (zone1-current) but don't mind those. > > 5- Move the zones to zfs on the 'stage' BE. > # zoneadm -z zone1 halt > # zoneadm -z zone1 move /zones/zone1 > > 6- Copy back the new BE (stage) to the original boot device (where current > BE was) > NOTE: This is an additional step to put back the 'stage' BE to the 'current' > BE slices. This step should be done after the upgrade is successful as it > will destroy the 'current' BE. > We need to do this step manually so Live Upgrade will not attempt to move > the zones to 'current' BE, we would want to keep the zones where they are on > zfs. > # newfs -m 1 /dev/dsk/c1t0d0s0 # this is root > # newfs -m 1 /dev/dsk/c1t0d0s3 # this is var > # mkdir /a > # mount /dev/dsk/c1t0d0s0 /a > # mkdir /a/var > # mount /dev/dsk/c1t0d0s3 /a/var > # ufsdump 0f - /dev/rdsk/c1t0d0s5 | (cd /a; ufsrestore xf -) > # installboot /usr/platform/`uname -i`/lib/fs/ufs/bootblk > /dev/rdsk/c1t0d0s0 > # vi /a/etc/vfstab # to modify the boot and target devices and > replace s5 with s0 and s3 (var). > # eeprom boot-device='disk0 /[EMAIL PROTECTED],600000/[EMAIL > PROTECTED]/[EMAIL PROTECTED]/[EMAIL PROTECTED]/[EMAIL PROTECTED]/[EMAIL > PROTECTED]/[EMAIL PROTECTED],0:f' # in case disk0 does not boot. > # init 6 > > You should be done at this point. > > > This message posted from opensolaris.org > _______________________________________________ > zfs-discuss mailing list > zfs-discuss@opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/zfs-discuss > -- Asif Iqbal PGP Key: 0xE62693C5 KeyServer: pgp.mit.edu _______________________________________________ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
