lvextend: Couldn't resynchronize logical volume

[Tonny Sejr Kromann]

Hi

I changed a defective non-boot disk, but afterwards I cannot mirror one of the 2 lvols:
lvextend -m 1 /dev/vg00/lvol8
The newly allocated mirrors are now being synchronized. This operation will
take some time. Please wait ....
lvextend: Couldn't re-synchronize stale partitions of the logical volume:
I/O error
lvextend: Couldn't resynchronize logical volume "/dev/vg00/lvol8".

What might the problem be?

--
Regards
Tonny

[Stefan Farrelly]

Either the source disk or the replacement disk is bad - generatnig the i/o error.

To find out which one it is do a non-destructive dd on each disk;

dd if=/dev/rdsk/cxxxxxx of=/dev/null bs=1024k

One of the dd's will fail with an i/o error, showing you which disk is bad and you will need to replace it.

[Ravi_8]

Hi,

can you do
#dd if=/dev/rdsk/cxtxdx of=/dev/null bs=1024k

(cxtxdx are root as well inserted disk)
you have to perform dd on bith the disks, if it yields any I/O error then disk is in problem.

if both disks are fine then mirroring should be ok.

[Jean-Louis Phelix]

Hi,

It could also be that LVM has "deactivated" the disk for this volume. After replacing it, reactivate the vg using -a y (or e if exclusive mode). From vgchange man pages :

If the -a y or -a e option is executed on a currently active volume group, vgchange attempts to include any physical volumes that were previously listed as missing. This is useful if a physical volume has come back online.
However, no automatic synchronization of any mirrored logical volumes is done. If synchronization is required, execute the vgsync command (see vgsync(1M)).

Regards.

[Cheryl Griffin]

Check if the extends are stale:
# lvdisplay -v /dev/vg00/lvol8 | grep stale

[S.K. Chan]

Typically in the replacement of a defective boot disk you would run ..
# pvcreate -B /dev/rdsk/cXtXdX
# mkboot -l /dev/rdsk/cXtXdX
# mkboot -a "hpux -lq" /dev/rdsk/cXtXdX
# vgcfgrestore -n vg00 /dev/rdsk/cXtXdX
# vgchange -a y vg00
# vgsync vg00
For a mirrored non-boot disk you would skip the mkboot steps. It does looks like a bad disk (like others have mentioned). You can also run STM exerciser tool on it to verify the disk. Also check in /var/adm/syslog/syslog.log for any IO or SCSI errors. If you cannot find anything wrong with the disk, it could be LVM problem (not patched with the latest patch ..) but it's very rare.