replace bootdisk

[Vlad_9]

Hi all!
I have two physical boot volumes in vg00.
Primary Boot /dev/dsk/c2t2d0 (0/0/2/0.2) and Mirror Boot /dev/dsk/c1t2d0 (0/0/1/1.2)
Primary bootdisk failed, i need replace it.

Sequence of actions:

Installed old disk.
1. Remove mirror:
# lvreduce -m 0 -A n -k /dev/vg00/lvol1 /dev/dsk/c2t2d0
- for all logical volumes (1-8), included root(3) and swap(2)

2. Remove disk from volume group:
# vgreduce /dev/vg00 /dev/dsk/c2t2d0

3.Turn off server, remove old disk and install new one.
Turn on server, it's boot from secondary disk.

4.Create physical volume PV:
#pvcreate -B -f /dev/rdsk/c2t2d0

5. Add disk into volume group:
#vgextend /dev/vg00 /dev/dsk/c2t2d0

6. Make this disk as boot:
#mkboot /dev/rdsk/c2t2d0
#mkboot -a "hpux /stand/vmunix" /dev/dsk/c2t2d0

6. Make mirror:
#lvextend -m 1 /dev/vg00/lvol1 /dev/dsk/c2t2d0
- for all logical volumes(1-8)

7. And after:

#lvlnboot -r /dev/vg00/lvol3 #root
#lvlnboot -s /dev/vg00/lvol2 #swap
#lvlnboot -b /dev/vg00/lvol1 #boot
#lvlnboot -d /dev/vg00/lvol2 #dump

Verify result of mirroring:
#lvlnboot -v


So, it's ok?
Have I not missed something?

[Luk Vandenbussche]

Hi Vlad,

Check this document

http://docs.hp.com/en/5991-1236/When_Good_Disks_Go_Bad.pdf

[Sandy Chen]

Hi Vlad,

Looks okay for me. May I know what type of machine you have? Is it necessary to turn off server? Most of the disk are hot swap nowadays.

Regards,
Sandy

[Vlad_9]

Sandy, it's 9000 series L1000 enterprise server solution.

Luk, thanks for link.

[Sandy Chen]

Okay then, you're right, you should turn off the server.

Good luck

regards,
Sandy

[Geoff Wild]

You don't need to lvreduce...

For a root disk:

After failed disk has been replaced by HP (assuming the disk is c1t5d0):

vgcfgrestore -n /dev/vg00 /dev/rdsk/c1t5d0

mkboot /dev/rdsk/c1t5d0

mkboot -a "hpux -lq (;0) /stand/vmunix" /dev/rdsk/c1t5d0

vgchange -a y /dev/vg00

lvlnboot -Rv /dev/vg00

vgsync /dev/vg00


That said, read the following:

1) If the disk is completely "dead", such as if you run ioscan and status is "no_hw" then you can hot swap the disk online.

2) However in circumstances where the disk has not fully failed please do one of the following to avoid data corruption :

a) reduce mirror before replacing the disk
b) deactivate VG before replacing the disk
c) shutdown system to replace the disk


Rgds...Geoff

[A. Clay Stephenson]

I would amend Geoff's statement slightly:

0) Confirm that all extents are current on the remaining "good" disk.

1a) If the disk is completely "dead", such as if you run ioscan and status is "no_hw" then you can hot swap the disk online.

1b) If the disk is not completely dead, convert it to completely dead status by sliding the disk module out a few centimeters and allowing it to spin down. Wait one minute or so so that the failure is detected.

I've used this method for well over ten years and have never shutdown to replace a disk (boot disk or otherwise) and have never had a problem --- but maybe The Force is with me. During that time I have replaced many tens of disks.

[Vlad_9]

Hi Geoff!

Unfortunately, i run lvreduce for lvol1 and
get this output:
'logical volume "/dev/vg00/lvol1" has been successfully reduced'
and
lvlnboot: logical volume has no extents

Then i stopped...
How i can correct this?

[Vlad_9]

Thanx to all.

vgcfgrestore should be added to my list of commands.

Geoff, thank you for advice.

[Marcel Burggraeve]

Might be a bit late but personally I would never ( if possible at least ) turn off a server when one disk of a mirror has failed.
If you're unlucky you'll end up with the other disk failing as well because it doesn't wanna spin up anymore or start working properly after it has been spin down.