# zpool status
pool: rpool
state: ONLINE
scrub: none requested
config:

NAME        STATE     READ WRITE CKSUM
rpool       ONLINE       0     0     0
c0t0d0s0  ONLINE       0     0     0

errors: No known data errors

Now verify that the two disks that you want to mirror looks the same at the slave level.

# format
Searching for disks…done

AVAILABLE DISK SELECTIONS:
0. c0t0d0 <ST320011A cyl 38790 alt 2 hd 16 sec 63>
/pci@1f,0/ide@d/dad@0,0
1. c0t2d0 <DEFAULT cyl 38790 alt 2 hd 16 sec 63>
/pci@1f,0/ide@d/dad@2,0
Specify disk (enter its number): 0
selecting c0t0d0
[disk formatted, no defect list found]
/dev/dsk/c0t0d0s0 is part of active ZFS pool rpool. Please see zpool(1M).
/dev/dsk/c0t0d0s2 is part of active ZFS pool rpool. Please see zpool(1M).

FORMAT MENU:
disk       – select a disk
type       – select (define) a disk type
partition  – select (define) a partition table
current    – describe the current disk
format     – format and analyze the disk
repair     – repair a defective sector
show       – translate a disk address
label      – write label to the disk
analyze    – surface analysis
defect     – defect list management
backup     – search for backup labels
verify     – read and display labels
save       – save new disk/partition definitions
volname    – set 8-character volume name
!<cmd>     – execute <cmd>, then return
quit
format> ver

Primary label contents:

Volume name = <        >
ascii name  = <ST320011A cyl 38790 alt 2 hd 16 sec 63>
pcyl        = 38792
ncyl        = 38790
acyl        =    2
nhead       =   16
nsect       =   63
Part      Tag    Flag     Cylinders         Size            Blocks
0       root    wm       0 – 38789       18.64GB    (38790/0/0) 39100320
1 unassigned    wm       0                0         (0/0/0)            0
2     backup    wm       0 – 38789       18.64GB    (38790/0/0) 39100320
3 unassigned    wm       0                0         (0/0/0)            0
4 unassigned    wm       0                0         (0/0/0)            0
5 unassigned    wm       0                0         (0/0/0)            0
6 unassigned    wm       0                0         (0/0/0)            0
7 unassigned    wm       0                0         (0/0/0)            0

format> disk 1
selecting c0t2d0
[disk formatted, no defect list found]
format>

Notice that both disks have 0 – 38789 Cylinders.  So the disks look good, now we can attach the second disk.

# zpool attach rpool c0t0d0s0 c0t2d0s0
#
# zpool status
pool: rpool
state: ONLINE
status: One or more devices is currently being resilvered.  The pool will
continue to function, possibly in a degraded state.
action: Wait for the resilver to complete.
scrub: resilver in progress for 0h0m, 3.28% done, 0h24m to go
config:

NAME          STATE     READ WRITE CKSUM
rpool         ONLINE       0     0     0
mirror      ONLINE       0     0     0
c0t0d0s0  ONLINE       0     0     0
c0t2d0s0  ONLINE       0     0     0

errors: No known data errors

WoW that was easy to mirror two disks with ZFS.  No need to set up data stores or anything like that.  The fact that the zpool status output specify “mirror” means the disks are mirrored and not striped.  If you have multiple zfs pools you can specify zpool status POOLNAME.

Now to try to distroy the disk, ha, ha, ha!

# dd if=/dev/random of=/dev/rdsk/c0t2d0s0
969+0 records in
969+0 records out

Now if that disk was on a normal RAID utility stuch as solstice disk suite SDS or even hardware raid then both disks would be destroyed because they operate at the block level, and will copy the random blocks back do c0t0d0s0

# zpool status
pool: rpool
state: DEGRADED
status: One or more devices could not be opened.  Sufficient replicas exist for
the pool to continue functioning in a degraded state.
action: Attach the missing device and online it using ‘zpool online’.
see: http://www.sun.com/msg/ZFS-8000-2Q
scrub: resilver in progress for 0h1m, 10.52% done, 0h12m to go
config:

NAME          STATE     READ WRITE CKSUM
rpool         DEGRADED     0     0     0
mirror      DEGRADED     0     0     0
c0t0d0s0  ONLINE       0     0     0
c0t2d0s0  UNAVAIL      0     0     0  cannot open

errors: No known data errors

Oha happy ZFS is already working on fixing or resilvering the disk that was written to with random data.

# zpool status
pool: rpool
state: DEGRADED
status: One or more devices could not be opened.  Sufficient replicas exist for
the pool to continue functioning in a degraded state.
action: Attach the missing device and online it using ‘zpool online’.
see: http://www.sun.com/msg/ZFS-8000-2Q
scrub: resilver completed after 0h0m with 0 errors on Tue Apr 14 12:55:48 2009
config:

NAME          STATE     READ WRITE CKSUM
rpool         DEGRADED     0     0     0
mirror      DEGRADED     0     0     0
c0t0d0s0  ONLINE       0     0     0
c0t2d0s0  UNAVAIL      0     0     0  cannot open

errors: No known data errors

Now ZFS is angry because we wrote random data to the disk.

# zpool online  rpool  c0t2d0s0

# installboot -F zfs /usr/platform/`uname -i`/lib/fs/zfs/bootblk /dev/rdsk/c0t2d0s0

ex. metadetach -f d0 d20
d0: submirror d20 is detached

2. Clear each submirror

ex. metaclear d20
d20: Concat/Stripe is cleared

3. Delete the db replica on the failed drive (you can find the replica output by issuing metadb command with no arguments):

ex. metadb -d c0t2d0s7

4. Remove failed drive. Replace with new drive.

5. Run the format command and place the correct partition table, using the layout of the non-replaced drive.

6. re-create the metadb replica on the new disk

ex. metadb -a c0t2d0s7

7. Run metainit for each replaced submirror

ex. metainit d21 1 1 c0t2d0s1
d21: Concat/Stripe is setup

8. Run metattach for each submirror

ex. metattach d0 d20
d0: submirror d20 is attached

ex. metattach d1 d21
d1: submirror d21 is attached

/etc/rc2.d/S70ndd

1000 full duplex autoneg

#!/sbin/sh
#
PATH=/sbin:/usr/bin:/usr/sbin

echo “Configuring Network Interfaces”
interface=bge0
case “$1″ in
# Example setting Gigabit Ethernet to Auto Negotiate
‘start’)
ndd -set /dev/$interface adv_1000fdx_cap 1
ndd -set /dev/$interface adv_1000hdx_cap 0
ndd -set /dev/$interface adv_100fdx_cap 0
ndd -set /dev/$interface adv_100hdx_cap 0
ndd -set /dev/$interface adv_10fdx_cap 0
ndd -set /dev/$interface adv_10hdx_cap 0
ndd -set /dev/$interface adv_autoneg_cap 1
;;
esac
exit 0_cap 0

Excerpt from /var/adm/messages

6 server05 mount[4691]: [ID 612810 user.error] mount: /www: Device busy
Feb 5 07:41:19 server05 mount[4708]: [ID 612810 user.error] mount: /www: Device busy
Feb 5 07:41:43 server05 mount[4717]: [ID 612810 user.error] mount: /www: Device busy
Feb 5 07:43:30 server05 mount[4949]: [ID 612810 user.error] mount: /www: Device busy
Feb 5 07:49:55 server05 mount[5273]: [ID 612810 user.error] mount: /www: Device busy
Feb 5 07:53:39 server05 mount[5656]: [ID 612810 user.error] mount: /www: Device busy
Feb 5 07:54:31 server05 mount[5713]: [ID 612810 user.error] mount: /www: Device busy

use fuser to identify the process that is causing the trouble.

server05% fuser /www
/www: 4053c

server05% ps -ef | grep 4053
abe 4061 4053 0 07:27:20 pts/3 0:00 less access
abe 4053 4031 0 07:26:50 pts/3 0:00 bash
abe 4075 4053 0 07:28:05 pts/3 0:00 less access
abe 4067 4053 0 07:27:42 pts/3 0:01 less access
abe 4122 4053 0 07:28:44 pts/3 0:00 more
abe 4121 4053 0 07:28:44 pts/3 0:00 grep -i keep-a access
louis 6983 6950 0 08:14:57 pts/1 0:00 grep 4053

# dumpadm -d /dev/dsk/c0t0d0s1
Dump content: kernel pages
Dump device: /dev/dsk/c0t0d0s1 (dedicated)
Savecore directory: /var/crash/unknown
Savecore enabled: no
# dumpadm -y
Dump content: kernel pages
Dump device: /dev/dsk/c0t0d0s1 (dedicated)
Savecore directory: /var/crash/unknown
Savecore enabled: yes