cancel
Showing results for 
Search instead for 
Did you mean: 

Disk went bad

 
Gary_OO
Occasional Advisor

Disk went bad

i have a K360 server where a disk went bad in vg01
The machine is running, but I am unable to remove the disk from vg01...
It appears the disk is "busy"
There are mirrors to this disk, which of course are now stale.

I want to get rid of this disk from my vg01, but am unable..

Thanks for any help!
lvreduce doesnt work to break the mirrors....

# pvdisplay /dev/dsk/c0t5d0
pvdisplay: Warning: couldn't query physical volume "/dev/dsk/c0t5d0":
The specified path does not correspond to physical volume attached to
this volume group
pvdisplay: Warning: couldn't query all of the physical volumes.
pvdisplay: Couldn't retrieve the names of the physical volumes
belonging to volume group "/dev/vg01".
pvdisplay: Cannot display physical volume "/dev/dsk/c0t5d0".
3 REPLIES 3
Gary_OO
Occasional Advisor

Re: Disk went bad

also couldnt vgreduce the disk from vg01

# vgreduce /dev/vg01 /dev/dsk/c0t5d0
vgreduce: Couldn't query physical volume "/dev/dsk/c0t5d0":
The specified path does not correspond to physical volume attached to
this volume group
Gary_OO
Occasional Advisor

Re: Disk went bad

tried to do a vgscan and it was ok for vg00 but nothing for vg01 because of the bad disk...
Did it in preview mode, same result without the -p option when i removed lvmtab...


# vgscan -p -v -a
vgscan: The physical volume "/dev/dsk/c0t3d0" is already recorded in the "/etc/l
vmtab" file.
vgscan: The physical volume "/dev/dsk/c0t4d0" is already recorded in the "/etc/l
vmtab" file.
vgscan: The physical volume "/dev/dsk/c0t5d0" is already recorded in the "/etc/l
vmtab" file.
vgscan: The physical volume "/dev/dsk/c0t6d0" is already recorded in the "/etc/l
vmtab" file.
Couldn't stat physical volume "/dev/dsk/c0t5d0":
Invalid argument
Couldn't stat physical volume "/dev/dsk/c1t2d0":
Invalid argument

vgscan: has no correspoding valid raw device file under /dev/rdsk.
Verification of unique LVM disk id on each disk in the volume group
/dev/vg01 failed.


/dev/vg00
/dev/dsk/c0t4d0
/dev/dsk/c0t6d0
Patrick Wallek
Honored Contributor

Re: Disk went bad

If you want to just remove the disk without replacing it, do the following:

 

1) Put the old /etc/lvmtab back into place.

 

2) You will need to reduce the mirrors off of the bad disk by using the '-k' option to lvreducde. 

For example: lvreduce -m 0 -k /dev/vg01/lvol1 /dev/dsk/c0t5d0

Do this for all LVs that are mirrored to this drive.

 

3) Reduce the disk out of VG01 with the '-f' option -- vgreduce -f /dev/vg01