lvextend -m error message

[Wenth Leopold]

HALLO TO EVERYONE
I GOT BELOW SOME ERROR MESSAGES
i dont know how to treat that message
please some comments would be appreciated

wdash1,sys,root # pvcreate -f /dev/rdsk/c0t8d0
wdash1,sys,root # pvcreate -B -f /dev/rdsk/c0t8d0

wdash1,sys,root # mkboot /dev/rdsk/c0t8d0
wdash1,sys,root # mkboot -a "hpux (;0)/stand/vmunix" /dev/rdsk/c0t8d0

wdash1,sys,root # vgextend /dev/vg00 /dev/dsk/c0t8d0
wdash1,sys,root # lvextend -m 1 /dev/vg00/lvol1 /dev/dsk/c0t8d0
The newly allocated mirrors are now being synchronized. This operation will
take some time. Please wait ....
Logical volume "/dev/vg00/lvol1" has been successfully extended.
vgcfgbackup: /etc/lvmtab is out of date with the running kernel:Kernel indicates
3 disks for "/dev/vg00"; /etc/lvmtab has 2 disks.
Cannot proceed with backup.
wdash1,sys,root #

[Karthik S S · Accepted solution]

Hi,

try these steps,

cd /etc
mv lvmtab lvmtab.org
vgscan
strings lvmtab

make sure that u have 3 disks now.

vgcfgbackup vg00

Regards,
Karthik

[Massimo Bianchi]

Hi,
checking with "strings /etc/lvmtab", do you see all your disks ?


Maybe your lvmtab is corrupt, you can think of re-creating it with the command "vgscan".

It would be better prior to check it with

"vgscan -p" that will be only a preview.

For a complete procedure:

cp -p /etc/lvmtab /etc/lvmtab.copy
vgscan -p
and see the output. if the output is correct, with teh two disks

mv /etc/lvmtab /etc/lvmtab.orig
vgscan

and wait...

HTH,
Massimo

[T G Manikandan]

check whether the CurPV & the Act PV disagree in the vgdisplay -v /dev/vg00 output/.


This doc can help u

http://www1.itrc.hp.com/service/cki/docDisplay.do?docLocale=en_US&docId=200000065011231

[Jon Mattatall]

This looks like something we saw here a few weeks ago....

Were you replacing a failed boot disk, and did you let the box come up normally instead of going to single user or maintenance mode?

What is the output of lvdisplay -vk /dev/vg00/lvol* ?

[Wenth Leopold]

hello to all of you
thanks for info
i think we found the reason
because we first replaced the defective disk anf THEN "killed" the mirrror

could that??be solution ??

[Massimo Bianchi]

Yes!
Because some of the data wa already written in the disks, althought not in the lvmtab.

Do a "vgdisplay -v vg00", i think you will see some strange things.

In this case, the best is to reduce vg00 pulling out c0t8d0.
You may need to use the "lvreduce" with the "-k" option, besause the disk will apper as missing.

OR, if you can play with it a bit, use vgcfgrestore to restore the information to the disk c0t8d0.

If this is a sandbox, i will try options.. they can be usefull in future !

HTH,
Massimo

[S. Eerkes]

i had the same problem,

do the following steps

1) vgreduce -g /dev/vg00
2) mv /etc/lvmtab /etc/lvmtab.old
3) vgscan -a
4) vgcfgbackup vg00

note: "The vgreduce command with -f option removes all missing physical volume from the volume group."

[Elmar P. Kolkman]

It could very well be that the vgreduce won't work because the system thinks the physical volume is still in use, though stale. By using the pv_key values lvdisplay -vk gives you can remove those:

lvreduce -k -m 0

Good luck.