problems with vgscan and vgdisplay on vg00

[R.L]

Hi all,

When I'm doing vgscan, I got this msg:
"Couldn't stat physical volume "/dev/dsk/c0t0d0": Invalid argument.
.....
The colume group /dev/vg00/group was not matched with any Physical Volumes."

And when I do #vgdisplay /dev/vg00, I got error msg:
"Volume group "/dev/vg00" does not exist in the "/etc/lvmtab" file.

Could anybody help me to solve this?

[Marc Dijkstra]

Have a look at the /etc/lvmconf/vg00.conf file (you will need to strings it) and see which disks are in that VG. As vg00 is generally your boot disk, I find it strange that you get these responses. When you do a bdf, what is the output WRT the / ?

How many disks are in your system? ioscan -fknC disk, and is /dev/dsk/c0t0d0 your boot disk?

MND

[Dan Hetzel]

Hi,

Make sure you have all the special files with 'insf -e'.

Try recreating /etc/lvmtab with 'vgscan -a'. Is this command showing the same error message?
Read the manpages of 'vgscan' as there could be some issues with alternate links.

If you are running 11.0:
Patch PHKL_21607 fixes the inability of HPUX 11.0 to support devices that have skipped LUN IDs.
It's been superseeded by PHKL_22759
ftp://ftp.itrc.hp.com/hp-ux_patches/s700_800/11.X/PHKL_22759

Best regards,

Dan

[R.L]

More information and background of my problem:
I run "umount -a" when I'm in single user mode.
Then my problems arise. I cannot do mount anymore. As a result, most of the command cannot be used.

I got 2 disks in my system and c0t0d0 is not my boot disk. vgscan -a gave me the same error messages.

[Marc Dijkstra]

Is /dev/vg00 striped across the two disks in the system? as you say that /dev/dsk/c0t0d0 is not your boot disk, why is vgscan/vgdisplay complaining about the disk wrt /dev/vg00?

what is the output of the /etc/lvmtab file (use strings) and /etc/fstab files?

MND

[R.L]

Marc,

I'm not sure if /dev/vg00 striped across the 2 disks and I can't issue the strings command.

[Marc Dijkstra]

Can you cat the file? As the cat command (and lv/vg commands sit in the /sbin directory on the root)

How about rebooting the machine into maintenance mode and seeing if any of the commands work? Reboot the machine and catch the boot sequence, then issue the boot pri ipl command and get to the secondary loader (ISL) then issue hpux -lm and see if anything works.

MND