n4000 mirror boot disk

[irfan shahmiri]

Hi there

We have an HP N4000. The "uname -a" gives
HP-UX ux01pwow B.11.00 A 9000/800 658349303 two-user license.

It has 2 internal disk and 2 SC10(not sure of the name) disk arrays with 10 disks each (18g).

Currently the boot disk is not mirrored. The two internal 9G disks are in the vg00 VG.

We are trying to come up with a DR plan for this system and I have a few questions. We have created a tape using make_tape_recovery -A.

1. If we lose one disk from the vg00 VG, can we add one disk from the external disk array to replace the the internal disk while using the tape to recover?

2. Has anyone done a recovery using the tape? Are there any best practices out there for tape recovery?

thanks

[Paul Eadington_1]

To answer the questions.

1. I think you could, there would be some manual tasks but as long as you went for an interactive recovery process then you should be able to redefine vg00 from the GUI to include the disk from the array.

2. Yes I have and it was a painful process, mainly as the make_recovery_tape we had was created with the -A option. This only brings back what the system defines as essential for the OS and you will have to do a lot of recovery work from backup afterward to finish the task.

I'd advise creating your make recovery tape with make_tape_recovery -x inc_entire=vg00 which will include all data from vg00 instead of just the essential files.

The list of essential areas can be found in /opt/ingite/recovery/mnr_essentials.

A qusetion from me, do you use both the internals for data? If you pvdisplay -v /dev/dsk/c?t?d? are extents from both disks being used or does one have all free?

[Steven E. Protter]

pvcreate -B /dev/rdsk/c1t0d0 #use real disk

mkboot -l /dev/rdsk/c1t0d0
mkboot -a "hpux -lq (;0)/stand/vmunix" /dev/rdsk/c1t0d0 # use real disk


# mkboot -b /usr/sbin/diag/lif/updatediaglif -p ISL -p AUTO -p HPUX -p PAD -p LABEL /dev/rdsk/c?t?d?

If you are running 64-bit OS:

# mkboot -b /usr/sbin/diag/lif/updatediaglif2 -p ISL -p AUTO -p HPUX -p PAD -p LABEL /dev/rdsk/c?t?d?


vgextend /dev/vg00 /dev/dsk/c1t0d0 # same thing
lvextend -m 1 /dev/vg00/lvol1 /dev/dsk/c1t0d0

# real disk. repeat for other lvols

lvlnboot -r /dev/vg00/lvol3 # root fs /
lvlnboot -s /dev/vg00/lvol2 #swap
lvlnboot -d /dev/vg00/lvol2 #swap/dump
lvlnboot -b /dev/vg00/lvol1
lvlnboot -R
lvlnboot -v
setboot
setboot -a 52.1.0 # second disk

Best DR idea is regular make_tape_recovery tapes rotated off site.

Do see about arranging for similar hardware to be availablein the DR center.

The make_tape_recovery will recover to similar hardware but not radically differeent hardware.

SEP

[Alexander M. Ermes]

Hi there.
1.
Yes, you can. We have done this on two N4000
setup as cluster and it works fine.

2.
Yes, we had to do this after a mainboard failure on one of the cluster nodes.
REMARK :
Do a cpio or tar backup of your complete backup software from time to time ( if you
use something else then fbackup ).

Make a recovery tape every week.

You have to restore some directories with your normal backup software after you did the restore from the recovery tape to avoid old versions of something. Be very careful about this.

Rgds
Alexander M. Ermes

[Paul Torp]

i have made a boot-disk script that i always use:

#SYNTAX ./mirroring.sh cxtxdx (Verified unused disk)
#mirroring.sh:
------------
integer BITS
pvcreate -B /dev/rdsk/$1 # or pvcreate -fB /dev/rdsk/$1
vgextend /dev/vg00 /dev/dsk/$1
mkboot -l /dev/rdsk/$1
mkboot -a "hpux -lq" /dev/rdsk/$1

for x in $(ls -al /dev/vg00/lv* | sort +5 | awk '{print $10}')
do
lvextend -m 1 ${x} /dev/dsk/$1
done
lvlnboot -r /dev/vg00/lvol3
lvlnboot -b /dev/vg00/lvol1
lvlnboot -s /dev/vg00/lvol2
lvlnboot -d /dev/vg00/lvol2
lvlnboot -v

lifls -l /dev/rdsk/$1


BITS=$(/bin/getconf KERNEL_BITS)

if [ $BITS = 64 ]
then
mkboot -b /usr/sbin/diag/lif/updatediaglif2 -p ISL -p AUTO -p HPUX -p PAD -p LABEL /dev/rdsk/$1
else
mkboot -b /usr/sbin/diag/lif/updatediaglif -p ISL -p AUTO -p HPUX -p PAD -p LABEL /dev/rdsk/$1
fi



After that u manually add the disk in /stand/bootconf

hope this helps u out!

-Paul

[stone_wei]

hi

to the second question. From my expirience, I fail to get the avaible disk from a FC60 to the N4000 server, there is different inferface between those disks.

[Saurav_1]

Confirm, I think the VG00 is used to mirror both the Boot disks. So you should boot from the other disk in case one disk in VG00 fails.

make_tape_recovery -Ava /dev/rmt/0mn

Remember it will not retrive your mirroring during restoration. Check man page of the same.

[G. Vrijhoeven]

Hi,

I only like to add one reminder.
You could use the external disks if the size of the disks do not exceed the MAX PE * PE SIZE limit of the volumegroup. In other words if the disk size of the external disks does not exceed the size of the first disk that has been put in the volumegroup. ( or the size that has been adjusted by the vgcreate -e command)

Regards,

Gideon