need assistance on the disksuite front?

Just interested in mirroring my boot and swap.  :)

want steps?

absolutely.

are disks partitioned identically? in format, view partition table of disk 0,
<na>me it, go choose disk 1, view partition table, <se>lect whatever you named
the partition table of 0, <la>bel it.
(or do it with prtvtoc and fmthard, but be CAREFUL:
  shale% sudo prtvtoc /dev/rdsk/c1t0d0s2 | sudo fmthard -s - /dev/rdsk/c1t1d0s2
)


okay. now:

/usr/sbin/metadb -a -f -c3 -l500 /dev/rdsk/c0t0d0s7 \
	/dev/rdsk/c0t1d0s7 

(if your 5mb metadb partition is s7). -a = add; -f = force
c3 -l500 = 3 metadb's per metadb partition, 500 bytes per metadb.

if there is a third drive you have connected to the system it would be VERY
advisable to have a 5mb partition on it as well with metadb's. that can always
be done later, though.

why 3 metadbs?

in the event of disk failure, disksuite wants >50% of the metadb
partitions available. if one disk fails, you don't have >50%. Of course you
can always boot off of cd, but if you have three, you're guaranteed to always
have >50% in the event of a disk failure.

put the following in /etc/lvm/md.tab:

d0 -m d1
#d0 -m d1 d2
d1 1 1 c0t0d0s0
d2 1 1 c0t1d0s0

d3 -m d4
#d3 -m d4 d5
d4 1 1 c0t0d0s1
d5 1 1 c0t1d0s1

8<------8<
yes, the #'ed out lines are important, you'll see why in a sec.


run:

/usr/sbin/metainit -a -f
a = all in md.tab that aren't already set up
f = force -- it'll bitch about the fact that you're metainit'ing a live root
and swap device, that'll shut it up.

/usr/sbin/metaroot d0
("patches location in kernel" and updates /etc/vfstab)

next: vi /etc/vfstab. change your swap device to /dev/md/dsk/d3 and note how
your root device was automagically changed for you by metaroot

reboot.

df -k and swap -l to verify you're running on the metadevices.

yes.  should see higher performance now, too?

once we finish, yeah.
r/w should be round-robined. minor performance enhancement.
now, /usr/sbin/metattach d0 d2 and /usr/sbin/metattach d3 d5
attach the second components to the mirrors.
[listen to drives start chugging as they sync]

modify /etc/lvm/md.tab; comment out the existing d0 and d3 lines;
uncomment the other d0 and d3 lines.

you can watch status and sync stuff with /usr/sbin/metastat | more
or, even better, | grep 'done'

congratulations, you have just mirrored your OS and swap disks.

only reason for the reboot is if you sync the drives before, you get confetti.
meaning, it'll sync the data on the two drives, not knowing which one actually
had the data you wanted to, uh, keep.

--------------------------------------------------------------------
Date: Thu, 05 Feb 2004 07:45:28 -0600
From: J.D. Bronson <jd.bronson@wixb.com>
To: berry@housebsd.org
Subject: a little more


The system starts booting off of sds-mirror. However, because there are
only two of the original four state database replicas available, a quorum
is not achieved. The system requires manual intervention to remove the two
failed state database replicas:

Starting with DiskSuite 4.2.1, an optional /etc/system parameter exists
which allows DiskSuite to boot with just 50% of the state database replicas
on-line. For example, if one of the two boot disks were to fail, just two
of the four state database replicas would be available. Without this
/etc/system parameter (or with older versions of DiskSuite), the system
would complain of "insufficient state database replicas", and manual
intervention would be required on boot up. To enable the "50% boot"
behavior with DiskSuite 4.2.1, execute the following command:


# echo "set md:mirrored_root_flag=1" >> /etc/system





--
J.D. Bronson - "LoneBandit"
Aurora Health Care // Information Services // Milwaukee, WI USA
Office: 414.978.8282 // Email: jd@aurora.org // Pager: 414.314.8282