topic Re: 1 stale extent in root volume in Operating System - HP-UX

1 stale extent in root volume

Dispatching — Mon, 18 May 2009 10:47:56 GMT

I have 1 stale extent in /dev/vg00/lvol3 on disk c2t2d0.
This disk is replaced already twice without success. Lvsync doesn't work. Diskinfo and dd don't show (expected) errors. What can I do to fix this? The system doesn't boot from this disk anymore and I'm afraid when I'd try to break the mirror and recreate it, this will cause problems.
lvsync /dev/vg00/lvol3
lvsync: Couldn't re-synchronize stale partitions of the logical volume:
I/O error
lvsync: Couldn't resynchronize logical volume "/dev/vg00/lvol3".

#dd if=/dev/dsk/c1t2d0 of=/dev/null bs=1024k
346+0 records in
346+0 records out
root@cdrc4:/root#dd if=/dev/rdsk/c1t2d0 of=/dev/null bs=1024k
17366+1 records in
17366+1 records out
#dd if=/dev/rdsk/c2t2d0 of=/dev/null bs=1024k
17366+1 records in
17366+1 records out

root@cdrc4:/root#diskinfo -v /dev/dsk/c2t2d0
diskinfo: Character device required
root@cdrc4:/root#diskinfo -v /dev/rdsk/c2t2d0
SCSI describe of /dev/rdsk/c2t2d0:
vendor: HP 18.2G
product id: MAN3184MC
type: direct access
size: 17783240 Kbytes
bytes per sector: 512
rev level: HP04
blocks per disk: 35566480
ISO version: 0
ECMA version: 0
ANSI version: 2
removable media: no
response format: 2
(Additional inquiry bytes: (32)4e (33)30 (34)31 (35)30 (36)34 (37)32 (38)32 (39)20 (40)20 (41)20 (42)20 (43)46 (44)55 (45)4a (46)49 (47)54 (48)53 (49)55 (50)20 (51)0 (52)0 (53)0 (54)40 (55)b (56)18 (57)1 (58)3c (59)1 (60)9b (61)0 (62)0 (63)0 (64)0 (65)0 (66)0 (67)0 (68)0 (69)0 (70)0 (71)0 (72)0 (73)0 (74)0 (75)0 (76)0 (77)0 (78)0 (79)0 (80)0 (81)0 (82)0 (83)0 (84)0 (85)0 (86)0 (87)0 (88)0 (89)0 (90)0 )
root@cdrc4:/root#diskinfo -v /dev/rdsk/c1t2d0
SCSI describe of /dev/rdsk/c1t2d0:
vendor: HP 18.2G
product id: MAN3184MC
type: direct access
size: 17783240 Kbytes
bytes per sector: 512
rev level: HP04
blocks per disk: 35566480
ISO version: 0
ECMA version: 0
ANSI version: 2
removable media: no
response format: 2
(Additional inquiry bytes: (32)4e (33)30 (34)30 (35)31 (36)34 (37)31 (38)32 (39)20 (40)20 (41)20 (42)20 (43)46 (44)55 (45)4a (46)49 (47)54 (48)53 (49)55 (50)20 (51)0 (52)0 (53)0 (54)40 (55)b (56)18 (57)1 (58)3c (59)1 (60)9b (61)0 (62)0 (63)0 (64)0 (65)0 (66)0 (67)0 (68)0 (69)0 (70)0 (71)0 (72)0 (73)0 (74)0 (75)0 (76)0 (77)0 (78)0 (79)0 (80)0 (81)0 (82)0 (83)0 (84)0 (85)0 (86)0 (87)0 (88)0 (89)0 (90)0 )

Re: 1 stale extent in root volume

Jozef_Novak — Mon, 18 May 2009 10:55:03 GMT

Hello,

have you followed the correct procedure for replacing disks that are under LVM control ?

http://docs.hp.com/en/5991-1236/When_Good_Disks_Go_Bad_WP.pdf

J.

Re: 1 stale extent in root volume

Steven E. Protter — Mon, 18 May 2009 11:00:19 GMT

Shalom,

Break the mirror on all logical volumes with lvsplit or lvextend -m 0

re-mirror your system:
http://www.hpux.ws/?p=21

SEP

Re: 1 stale extent in root volume

Ganesan R — Mon, 18 May 2009 11:11:07 GMT

Hi,

I suspect that your primary disks is having statle extends. It is not uncommon that lvdisplay still shows synced but the actual extends has I/O error.

So the only way is break the mirror and check with lvsync on primary then add the secondary disk and extend the mirror. But take the ignite backup before you try anything. Ignite backup may fail if primary copy of lvol3 is having physical error

Be prepared to have recovery from ignite in case of failure.

Re: 1 stale extent in root volume

R.K. # — Mon, 18 May 2009 12:20:56 GMT

HEllo,

Can you find any stale extents in lvol3:
# lvdisplay -v /dev/vg00/lvol3 | grep -i stale

Which disk have stale extents?

Regds,
R.K.

Re: 1 stale extent in root volume

Dispatching — Wed, 27 May 2009 08:44:55 GMT

Hi Guys, thx for your replies, sorry for my late answer
I made a little error. The only stale extent is on my primary disk c1t2d0(lvol3)
00004 /dev/dsk/c1t2d0 00774 stale /dev/dsk/c2t2d0 00537 current
lvdisplay shows : available/stale not synced
Disk c1t2d0 has been replaced twice by our supplier...I suppose in a correct way.
I read the pdf "when good disks go bad" and that's why I already did the dd and diskinfo commands.
With no errors, which is strange because according the pdf there should be.
That's why I'm a bit suspicious. Still go for breaking/recreating mirror ?

Re: 1 stale extent in root volume

R.K. # — Wed, 27 May 2009 08:52:33 GMT

Hello,

Breaking and re-mirroring, is never an issue.
You can go ahead and try this.

You can also try "lvsync /dev/vg00/lvol3"

Regds,
R.K.

Re: 1 stale extent in root volume

johnsonpk — Wed, 27 May 2009 09:01:12 GMT

Hi,

00004 /dev/dsk/c1t2d0 00774 stale /dev/dsk/c2t2d0 00537 current

remove the mirror form primary disk and create a mirror again

#lvreduce -m 0 /dev/vg00/lvol3 /dev/dsk/c1t2d0

#lvextend -m 1 /dev/vg00/lvol3 /dev/dsk/c1t2d0

if it is not succeeded you may need to replace the disk again.

Thanks!!
Johnson

Re: 1 stale extent in root volume

Dispatching — Wed, 27 May 2009 11:43:42 GMT

I recreated the mirror on lvol3 without success. There is the exact same exent that remains stale.
#lvreduce -m 0 /dev/vg00/lvol3 /dev/dsk/c1t2d0
#lvextend -m 1 /dev/vg00/lvol3 /dev/dsk/c1t2d0
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/lvol3".
==>
00004 /dev/dsk/c2t2d0 00537 current /dev/dsk/c1t2d0 00774 stale

I will have the disk replaced for a 3th time.(a bad lot of disks last time maybe)

Re: 1 stale extent in root volume

Michal Kapalka (mikap) — Wed, 27 May 2009 11:56:15 GMT

Hi Dispatching,

sometimes, it is not a disk error but on the disk bay or SCSI card.

mikap

Re: 1 stale extent in root volume

johnsonpk — Wed, 27 May 2009 12:07:04 GMT

Hi

Better Replace it :-)

Johnson

Re: 1 stale extent in root volume

Patrick Wallek — Wed, 27 May 2009 12:07:31 GMT

I would definitely try replacing the disk again.

If it was any other component (cable, disk bay, scsi card) I would expect the entire disk to show bad repeatedly and not just one extent on the disk.

Re: 1 stale extent in root volume

Michal Kapalka (mikap) — Wed, 27 May 2009 12:19:19 GMT

Hi Dispatching,

could you please send a output in a file attachemnt from this commands ?

vgdisplay -v /dev/xyz

lvdisplay -v /dev/xyz/lvol(stale)

/etc/lvmtab

mikap

Re: 1 stale extent in root volume

Shailendran V Naidu — Wed, 27 May 2009 12:31:14 GMT

Hi,
I think there could be more stale extents in the disk - /dev/dsk/c1t2d0.

You can run a :
pvdisplay /dev/dsk/c1t2d0

and check the number of stale extents. and cross verify the same with

pvdisplay /dev/dsk/c2t2d0

If there are stale extents on both the disks then I am afraid the best thing to do is replace both disk and reignite the box.
If one disk is an issue then it is definately a HDD problem. You may want to contact the response center and get it fixed.

Stale extent issue is mostly a HDD media failure.

You can also try
lvsync /dev/vg00/lvol3

Re: 1 stale extent in root volume

R.K. # — Fri, 29 May 2009 01:57:34 GMT

Hello,

00004 /dev/dsk/c2t2d0 00537 current /dev/dsk/c1t2d0 00774 stale <==

With the above line, we see that PE 00537 of disk1 is trying to get mirrored with PE 00774 of disk2.
This is a Vg00 disk so probably they should be mirorred with their respective extents.PE 00537 with PE 00537 and so on.

Example:
LE PV1 PE1 Status 1 PV2 PE2 Status 2
0000 /dev/dsk/c0t1d0 0000 current /dev/dsk/c0t2d0 0000 current
0001 /dev/dsk/c0t1d0 0001 current /dev/dsk/c0t2d0 0001 stale
0002 /dev/dsk/c0t1d0 0002 current /dev/dsk/c0t2d0 0002 current
0003 /dev/dsk/c0t1d0 0003 current /dev/dsk/c0t2d0 0003 current

NOTE: PE1 and PE2 are same.

So check if there is any issue in mirroring with lvols right from lvol1, although stale mostly refers to faulty disk, but you have replaced disk for the 3rd time. Still you can look around this one.

Regds,
R.K.