lvlnboot: Couldn't query physical volume "/dev/dsk/c1t2d0":

[Carme Torca]

Hi,

I have a problem with one disk. I have a vg00 with two disks, but one /dev/dsk/c1t2d0 is wrong???. When I do a ioscan, it is ok:
disk 0 0/0/1/1.2.0 sdisk CLAIMED DEVICE SEAGATE ST39204LC
/dev/dsk/c1t2d0 /dev/rdsk/c1t2d0

but when I do an lvlnboot -v, its show me:
> lvlnboot -v
lvlnboot: Couldn't query physical volume "/dev/dsk/c1t2d0":
The specified path does not correspond to physical volume attached to
this volume group

The same when I do an lvdisplay -v /dev/vg00/lvol1 (f.exemple):
--- Distribution of logical volume ---
PV Name LE on PV PE on PV
/dev/dsk/c2t2d0 21 21

--- Logical extents ---
LE PV1 PE1 Status 1 PV2 PE2 Status 2
00000 ??? 00000 stale /dev/dsk/c2t2d0 00000 current
00001 ??? 00001 stale /dev/dsk/c2t2d0 00001 current
00002 ??? 00002 stale /dev/dsk/c2t2d0 00002 current
00003 ??? 00003 stale /dev/dsk/c2t2d0 00003 current

What can I do???

Thanks!!!

[James R. Ferguson · Accepted solution]

Hi Carme:

It appears that you have a bad disk. Do:

# diskinfo -v /dev/rdsk/c1t2d0

If you can't query the disk, or you get a zero-size for the disk, then you need to replace the disk.

Regards!

...JRF...

[Shahul]

Hi

I think U will have to do vgcfgrestore in this HDD. Before that make sure that ioscan, diskinfo commands are working with this HDD. check /var/adm/syslog/syslog.log also for possible errors. What about pvdisplay, is it showing the PV details?

Best of luck
Shahul

[PIYUSH D. PATEL]

Hi,

Check with diskinfo and check whether any errors in syslog.log

Could be some loose connection with the disk or disk may be bad.

Piyush

[Carme Torca]

Correct!!
The disk has 0 Kbytes
Thanks very much!!!

[Robert Sipe]

Are you mirroring your disks in VG00? If so you should be able to replace the bad disk and resync. Read on. Sorry for the lousy formating.

To resilver a mirrored volume:

As root,

1.Execute ioscan -fn (or check the web page for the server) to determine the device files for the volume that needs resilvering. The device file will look
something like:

/dev/dsk/c0t#d0

Where: # is the SCSI interface target number for the disk.

2.If the volume you are resilvering is a boot volume:

Install the boot program on the disk:

mkboot -l /dev/dsk/c0t#d0

Where: # has the same meaning as above.

NOTE: The -l causes mkboot to treat the device as an LVM layout disk, regardless of whether or not it is currently set up as one.

Set the autoboot flag:

mkboot -a "hpux(;0)/stand/vmunix" /dev/rdsk/c0t#d0

Where: # has the same meaning as above.

3.Restore the volume group configuration:

vgcfgrestore -n /dev/vg00 /dev/rdsk/c0t#d0

Where: # has the same meaning as above.

NOTE: You may get a "Could Not Find Physical Volume" message. This is normal.

4.Reactivate the volume group:

vgchange -a y

You should see a "Successfully Changed" message.

5.View the physical extents and check for a status of stale:

lvdisplay /dev/vg00/lvol1

6.Resilver the disk mirror:

vgsync vg00



After You Finish

Verify the primary and alternate boot path:

setboot

The output should look similar to the following:

Primary bootpath : 10/0.6.0
Secondary bootpath : 10/12.5.0

Autoboot is ON (enabled)
Autosearch is ON (enabled)

Verify the LIF data:

lifcp /dev/dsk/c0t#d0:AUTO -

Where: # has the same meaning as above.

The output should look similar to the following:

hpux (;0)/stand/vmunix

If what you see is not correct, redo the mkboot -a step from above