Mirrored root disk failed but issues removing.

[TechC]

Hello,

I have an HP RP3440 running HPUX 11i V1.

I have a mirrored VG00 with a failed root disk. Usually when this happens I follow when good disks go bad and lvreduce the mirrors and remove the failed disk from the VG00 and then replace the disk and remirror. Usually this goes well.

This time I tried the same procedure with this.. for lvols 1-9

lvreduce -m 0 -A n -k /dev/vg00/lvol1-9

And I get this message for lvol 1 and 2.

Logical volume "/dev/vg00/lvol1" has been successfully reduced.
lvlnboot: Logical Volume has no extents.
lvlnboot: Logical Volume has no extents.

For lvol3 - 9 it just took 9 minutes to complete. I know that the reason is that the bad disk is still defined in lvmtab...

So obviously I am now worried.

Since this is an RP3440 the disks are hot pluggable which from what I read means that if the OS is not using the disk it is safe to remove, where as hot swappable means the the drive can be swapped regardless. 

I found this link that details the exact same issue but I can't tell what steps exactly were taken to remedy the situaion.

http://h30499.www3.hp.com/t5/System-Administration/what-if-hpux11-11-mirror-disk-is-out-and-want-to-lvreduce/td-p/4341288

So at this point I have removed my logical volumes from the bad disk but my bad disk is still a  member of VG00.

What steps can I perform to safely replace the root mirror disk and not lose data. Looking at the thread I posted above gives me an uneasy feeling.

What I was gonna do was.

1) vgreduce /vg00 /dev/dsk/c2t1d0

2) Swap in the new disk.

3) Remirror from scratch using When good disks go bad..

Can anyone lay out the safe plan?

Thanks

[Torsten.]

Have a look at

When_Good_Disks_Go_Bad_WP
http://bizsupport2.austin.hp.com/bc/docs/support/SupportManual/c01911837/c01911837.pdf

[TechC]

Hello,

I was able to vgreduce the bad disk from vg00 without issue..

After replacing the disk the system does not recognize the drive.. This is an HP 3440 running HPUX 11iv1. Can someone please share the command or procedure to make the system rescan the scsi bus and recognize the replacement disk?

I see scsimgr is under /usr/sbin, but there is no manpage on 11.11. I used it before on 11iv3 to replace the disk however I am not sure it works on 11.11.

Thanks

[Torsten.]

>> After replacing the disk the system does not recognize the drive.


Is it listed in ioscan or not?

[TechC · Accepted solution]

Forget it , i just rmsf -H path and then insf -H path and the disk is recognized.

Closing.