lvm not working - boot fails

[paul_xedos]

Hi,
This is a bit of an odd one which has me stumped. For some reason a working box now seems to have a broken lvm since a reboot. All I can do is log in single user where only / is mounted.

The box has one disk and ioscan sees the partitions:-

# ioscan -fnC disk
Class I H/W Path Driver S/W State H/W Type Description
=======================================================================
disk 0 0/0/0/0.0.0 sdisk CLAIMED DEVICE HP Virtual Disk
/dev/dsk/c0t0d0 /dev/rdsk/c0t0d0
/dev/dsk/c0t0d0s1 /dev/rdsk/c0t0d0s1
/dev/dsk/c0t0d0s2 /dev/rdsk/c0t0d0s2
/dev/dsk/c0t0d0s3 /dev/rdsk/c0t0d0s3


vgdisplay shows only one lvol (/dev/vg00/ISOLV) whatever that is. However according to /etc/fstab I should see the following:-

/dev/vg00/lvol3 / vxfs delaylog 0 1
/dev/vg00/lvol1 /stand vxfs tranflush 0 1
/dev/vg00/lvol4 /home vxfs delaylog 0 2
/dev/vg00/lvol5 /opt vxfs delaylog 0 2
/dev/vg00/lvol6 /tmp vxfs delaylog 0 2
/dev/vg00/lvol7 /usr vxfs delaylog 0 2
/dev/vg00/lvol8 /var vxfs delaylog 0 2

If I try and mount one of these it fails which is why the box is currently in single user:-
# mount /usr
mount: /dev/vg00/lvol7: No such file or directory

So far I've tried:-
* removing /etc/lvmtab and running vgscan -a
* running vgimport /dev/vg00 - returns "already exists in lvmtab"

As it's single user I can't run strings /etc/lvmtab to see how the file looks (strings is in /usr which doesn't mount

In short, I'm now at a loss on how to proceed

Thanks
Paul

[Torsten.]

Since this is a **virtual machine** you should check first if all the backing stores are available on the host and still assigned to the guest.

[paul_xedos]

Hi Torsten,
Thanks very much for replying. In the following excerpt hiptcd11 is the host box - hiptc055 the hpvm. I think it all checks out, though, as the disk being present on the host and being passed down to the vm

hiptcd11:/var/patch# hpvmstatus -P hiptc055
[Virtual Machine Details]
: : : :
[Storage Interface Details]
Guest Physical
Device Adaptor Bus Dev Ftn Tgt Lun Storage Device
====== ========== === === === === === ========= =========================
disk scsi 0 0 0 0 0 disk /dev/rdsk/c4t0d2

hiptcd11:/var/patch# ioscan -fnC disk
Class I H/W Path Driver S/W State H/W Type Description
===========================================================================

disk 2 0/3/1/0.1.16.0.0.0.1 sdisk CLAIMED DEVICE HP HSV200
/dev/dsk/c4t0d1 /dev/dsk/c4t0d1s2 /dev/rdsk/c4t0d1 /dev/rdsk/c4t0d1s2
/dev/dsk/c4t0d1s1 /dev/dsk/c4t0d1s3 /dev/rdsk/c4t0d1s1 /dev/rdsk/c4t0d1s3
disk 3 0/3/1/0.1.16.0.0.0.2 sdisk CLAIMED DEVICE HP HSV200
/dev/dsk/c4t0d2 /dev/dsk/c4t0d2s2 /dev/rdsk/c4t0d2 /dev/rdsk/c4t0d2s2
/dev/dsk/c4t0d2s1 /dev/dsk/c4t0d2s3 /dev/rdsk/c4t0d2s1 /dev/rdsk/c4t0d2s3

[Torsten.]

If the guest can go into single user, there is something present :-)


What HPVM version do you run?

Since the EVA is a multipathing device and you are using "raw" disks, any multipathing involved?


On the guest, what happens if you do

# vgdisplay -v vg00

[paul_xedos]

HPVM version could be a bit old :-

VM's Version Number : 1.20.0

No multi-pathing involved

vgdisplay is a bit odd as its reporting an lvol called ISOLV:-

# vgdisplay -v vg00
--- Volume groups ---
VG Name /dev/vg00
VG Write Access read/write
VG Status available
Max LV 255
Cur LV 11
Open LV 1
Max PV 16
Cur PV 1
Act PV 1
Max PE per PV 3153
VGDA 2
PE Size (Mbytes) 16
Total PE 3143
Alloc PE 2705
Free PE 438
Total PVG 0
Total Spare PVs 0
Total Spare PVs in use 0

--- Logical volumes ---
LV Name /dev/vg00/ISOLV
LV Status available/syncd
LV Size (Mbytes) 3104
Current LE 194
Allocated PE 194
Used PV 1


--- Physical volumes ---
PV Name /dev/dsk/c0t0d0s2
PV Status available
Total PE 3143
Free PE 438
Autoswitch On

[Torsten.]

This is the oldest and most unsupported version ever :-)

Consider to update!

However, multi pathed disks without multipathing is always a problem, but not now in this case.

Look:

Cur LV 11
Open LV 1

How about this:

# ll /dev/vg00

All device files for LVOLs listed (on guest)?

[paul_xedos]

I've now upgraded the hpvm code:-

hiptcd11:/# hpvmstatus -v
Version A.02.00.00

unfortunately, still no good.

ls -l of /dev/vg00 looks good though and suggests the volumes are there

hiptcd11:/# ll /dev/vg00
total 0
crw-r----- 1 root sys 64 0x000000 Nov 1 2006 group
brw-r----- 1 root sys 64 0x000001 Nov 1 2006 lvol1
brw-r----- 1 root sys 64 0x000002 Nov 1 2006 lvol2
brw-r----- 1 root sys 64 0x000003 Nov 1 2006 lvol3
brw-r----- 1 root sys 64 0x000004 Nov 1 2006 lvol4
brw-r----- 1 root sys 64 0x000005 Nov 1 2006 lvol5
brw-r----- 1 root sys 64 0x000006 Nov 1 2006 lvol6
brw-r----- 1 root sys 64 0x000007 Nov 1 2006 lvol7
brw-r----- 1 root sys 64 0x000008 Nov 1 2006 lvol8
crw-r----- 1 root sys 64 0x000001 Nov 1 2006 rlvol1
crw-r----- 1 root sys 64 0x000002 Nov 1 2006 rlvol2
crw-r----- 1 root sys 64 0x000003 Nov 1 2006 rlvol3
crw-r----- 1 root sys 64 0x000004 Nov 1 2006 rlvol4
crw-r----- 1 root sys 64 0x000005 Nov 1 2006 rlvol5
crw-r----- 1 root sys 64 0x000006 Nov 1 2006 rlvol6
crw-r----- 1 root sys 64 0x000007 Nov 1 2006 rlvol7
crw-r----- 1 root sys 64 0x000008 Nov 1 2006 rlvol8

[paul_xedos]

Please ignore the above ls listing as that is on the host when it should have been run on the guest. The guest looks like this:-

INIT: SINGLE USER MODE

INIT: Running /sbin/sh
# ls -l /dev/vg00
total 0
brw------- 1 root sys 64 0x00000b Apr 30 12:45 ISOLV
crw------- 1 root sys 64 0x000000 Apr 30 12:37 group
crw------- 1 root sys 64 0x00000b Apr 30 12:37 rISOLV

[Torsten. · Accepted solution]

All the device files are missing - this is the reason for your problem.

Try

# mknod /dev/vg00/lvol1 b 64 0x000001
# mknod /dev/vg00/rlvol1 c 64 0x000001
# mknod /dev/vg00/lvol2 b 64 0x000002
...
for all the lvols.

Then re-activate vg00 again

# vgchange -a y vg00
# mount -a

[paul_xedos]

Torsten,

Superb! Your last suggestion fixed it. Thanks very much for all your help. That's made my day

Paul