couldn't query physical volume
murator
I've following scenario:
vg00 with 4 disk ( i see in lvmtab )
vgdisplay -v vg00 give me :
vgdisplay: Warning: couldn't query physical volume "/dev/dsk/c2t2d0":
The specified path does not correspond to physical volume attached to
this volume group
vgdisplay: Warning: couldn't query physical volume "/dev/dsk/c1t0d0":
The specified path does not correspond to physical volume attached to
this volume group
vgdisplay: Warning: couldn't query physical volume "/dev/dsk/c2t0d0":
The specified path does not correspond to physical volume attached to
this volume group
vgdisplay: Warning: couldn't query all of the physical volumes.
--- Volume groups ---
VG Name /dev/vg00
VG Write Access read/write
VG Status available
Max LV 255
Cur LV 6
Open LV 7
Max PV 8
Cur PV 1
Act PV 1
Max PE per PV 10000
VGDA 2
PE Size (Mbytes) 8
Total PE 4340
Alloc PE 4340
Free PE 0
Total PVG 0
Total Spare PVs 0
Total Spare PVs in use 0
A diskinfo on disks not seen by lvm is ok.
ioscan is ok ( disks claimed )
dd from disks is ok.
Any idea on what to do or look for ?
HP_BRS
> VG Name /dev/vg00
> VG Write Access read/write
> VG Status available
> Max LV 255
> Cur LV 6
> Open LV 7
> Max PV 8
> Act PV 1 <---active number of disks in VG, same as above
> Max PE per PV 10000
> VGDA 2
> PE Size (Mbytes) 8
> Total PE 4340
> Alloc PE 4340
> Free PE 0
> Total PVG 0
> Total Spare PVs 0
> Total Spare PVs in use 0
>
If this were a situation of the disks actually being missing, Cur PV
would be four (assuming you thought this was a 4 disk volume group)
and Act PV would be one (indicating only one disk is active).
Essentially, the LVM header on this disk says that there is only one
disk is this volume group, and that is conflicting with information
that is either in the /etc/lvmtab or memory. Do a strings on
/etc/lvmtab and you will probably see all four disks. Also do a
strings on /etc/lvmconf/vg00.conf and see how how raw disks are
listed. Does this conflict with strings on
/etc/lvmconf/vg00.conf.old? Post the results please.
Let's read the header off of all four disks and see if they share the
correct VGID.
To get the VGID off of any disk, do the following:
# dd if=/dev/rdsk/cXtXdX of=/tmp/root.disk bs=8k skip=1 count=1
# od -Ax -tx /tmp/root.disk
This will give you some output like:
0000000 4c564d52 45433031 5629418b
3a7b0def
0000010 5629418b 3a7b0df5 3fff6c
0
0000020 d0 80 56
a
0000030 80 e8
8 e0
0000040 148 3ff000 400
16
0000050 1000 b6c 3ff400
1ff
0000060 0 0 0
0
*
0000400 44454645 43543031 0
0
0000410 0 0 0
0
*
0002000
On row 0000000, the last two fields are the machine ID and the PVID,
respectively. The machine ID should match the machine id provided by
`uname -a` when converted to decimal. The machine ID listed the is
the host that performed the pvcreate on this disk. For me:
# echo "0x5629418b=D" | adb
1445544331
# uname -a
HP-UX HP_BRS B.11.00 U 9000/869 1445544331 unlimited-user license
Now, on row 0000010, the first two fields are: The machine id (again)
and the VGID. All disks in the same volume group should share the
same VGID. For me, the volume group id is 3a7b0df5. Every disk in my
volume group should share that value. If they don't, then these disks
actually belong to another volume group.
HP BRS
murator
"HP_BRS" <born...@yahoo.com> wrote in message
news:bb82cc2.02031...@posting.google.com...
/869 1445544331 unlimited-user license
>
> Now, on row 0000010, the first two fields are: The machine id (again)
> and the VGID. All disks in the same volume group should share the
> same VGID. For me, the volume group id is 3a7b0df5. Every disk in my
> volume group should share that value. If they don't, then these disks
> actually belong to another volume group.
>
>
> HP BRS
Hello,
strings /etc/lvmtab gine me four disk for vg00
/dev/vg00
/dev/dsk/c1t2d0
/dev/dsk/c2t2d0
/dev/dsk/c1t0d0
/dev/dsk/c2t0d0
I haven't /etc/lvmconf/vg00.conf.old but only vg00.conf
strings give me
CONFIG01
/dev/vg00
/dev/rdsk/c1t2d0
/dev/rdsk/c2t2d0
/dev/rdsk/c1t0d0
/dev/rdsk/c2t0d0
LVMREC01
HPLVMBDR;x>
LVMREC01
;x1E
HPLVMBDR;x>
LVMREC01
LVMREC01
;x1E
~z{XVGDA0001
;x>z
VGSA0001;x>z
I've also cheched VGID following your procedure on each disk : is the same ,
and the same is machine ID.
Regards
HP_BRS
recover your system for worst case scenarios (ie Ignite). Also
consider calling in to the HP response center if this is a
production/mission critical system. There are details which have not
been discussed which may help explain how this volume group got into
this situation.
Use vgcfgrestore to restore the proper volume group header to each
physical volume - - - even the one disk that is currently active.
After doing this to all four PVs, re-run `vgchange -a y /dev/vg00` and
see if it now includes all four volumes.
# vgcfgrestore -n /dev/vg00 /dev/rdsk/c1t2d0 /* any errors? */
# vgcfgrestore -n /dev/vg00 /dev/rdsk/c2t2d0 /* any errors? */
# vgcfgrestore -n /dev/vg00 /dev/rdsk/c1t0d0 /* any errors? */
# vgcfgrestore -n /dev/vg00 /dev/rdsk/c2t0d0 /* any errors? */
# vgchange -a y /dev/vg00 /* Does it still complain about the 3
missing pvols? */
# vgdisplay -v /dev/vg00 /* Act PV and Cur PV should both == 4 */
"murator" <mur...@katamail.com> wrote in message news:<a6n4aq$ibq$1...@newsreader.mailgate.org>...
murator
> # vgcfgrestore -n /dev/vg00 /dev/rdsk/c2t2d0 /* any errors? */
> # vgcfgrestore -n /dev/vg00 /dev/rdsk/c1t0d0 /* any errors? */
> # vgcfgrestore -n /dev/vg00 /dev/rdsk/c2t0d0 /* any errors? */
> # vgchange -a y /dev/vg00 /* Does it still complain about the 3
> missing pvols? */
> # vgdisplay -v /dev/vg00 /* Act PV and Cur PV should both == 4 */
Hello,
doesn't work .
I've in planning , like suggest HP , shutdown and restart in maintenance
mode,
vgexport vg00 and vgimport it.
I believe is the only way.
Thanks
Regards