Only one LV extent is stale on a disk when lvdisplay a LVM

[Sammy_2]

When I do a lvdisplay on disk, only one logical extent appears to be stale. Rest of the LV extent show current status.
LV status is available/stale.
Is this a problem with the entire disk and it needs replacement or just one extent ? how would I fix that ?
Thanks
(see lvdisplay attach below)
=================================
# lvdisplay -v /dev/vgprod2/lvol8|pg
--- Logical volumes ---
LV Name /dev/vgprod2/lvol8
VG Name /dev/vgprod2
LV Permission read/write
LV Status available/stale
Mirror copies 1

--- Distribution of logical volume ---
PV Name LE on PV PE on PV
/dev/dsk/c2t11d0 1625 1625
/dev/dsk/c6t11d0 1625 1625
--- Logical extents ---
LE PV1 PE1 Status 1 PV2 PE2 Status 2
00000 /dev/dsk/c2t11d0 00000 current /dev/dsk/c6t11d0 00000 stale
00001 /dev/dsk/c2t11d0 00001 current /dev/dsk/c6t11d0 00001 current
00002 /dev/dsk/c2t11d0 00002 current /dev/dsk/c6t11d0 00002 current
00003 /dev/dsk/c2t11d0 00003 current /dev/dsk/c6t11d0 00003 current

[Patrick Wallek]

You can try doing an:

# lvsync /dev/vgprod2/lvol8

and see what happens. If it goes back to current and stays there, then you should be OK.

However, if it does not go to 'current' then your disk is probably on its way out and will need to be replaced soon.

[John Palmer]

What happens when you try to resync it?

lvsync /dev/vgprod2/lvol8

If that fails with an I/O error then I'd check the disk with:
dd if=/dev/rdsk/c6t11d0 of=/dev/null bs=64k

It's likely that you've got a single bad area on the disk, if you've got a support contract, get it replaced.

Regards,
John

[Sammy_2]

This is the message I get.

# lvsync /dev/dsk/c6t11d0
lvsync: Couldn't find the volume group to which
logical volume "/dev/dsk/c6t11d0" belongs.

[S.K. Chan · Accepted solution]

First use all the tips give to check c6t11d0 to really make sure it does not have any hardware problem. Also use STM (the exercise tool). If you can't find anything wrong with the hardware and all commands vgsync/lvsync does not help then my guess is the LVM header (VGRA / PVRA) of that disk (c6t11d0) may be messed up, hence it shows inconsistent info. If that's the case you can restore the header back from vgprod2.conf file (provided that fiel is the latest and it's good) .. like so ..
# vgcfgrestore -n vgprod2 /dev/rdsk/c6t11d0
then run ..
# vgchange -a y vgprod2
The vgchange activation will re-sync the VG.
Check it again if you still see the stale extent.

[Uday_S_Ankolekar]

The command you used is not correct

lvsync is works with lvname and not with the disk.
You need to use exact lvpath.
In our case try using
#lvsync /dev/vgprod2/lvol8

-USA..