re: LVM and no Quorum not present, or some physical volume(s) are missing.

[Gurcharan Sahota_2]

We have a RP3440 server running HP-UX 11i, and had to reboot after making a Database parameter change. [the database in question is UniVerse 10.1.8]

received the following error message;

vgchange: Couldn't activate volume group "/dev/vg01":
Quorum not present, or some physical volume(s) are missing.

LVM: VG 64 0x010000: Unable to initialize the latest configuration information.
LVM: VG 64 0x020000: Unable to initialize the latest configuration information.
vgchange: Couldn't activate volume group "/dev/vg02":
Quorum not present, or some physical volume(s) are missing.

vgdisplay: Volume group not activated.
vgdisplay: Cannot display volume group "/dev/vg01".
vgdisplay: Volume group not activated.
vgdisplay: Cannot display volume group "/dev/vg02".

How do i fix this?

[DCE]

It appears you have lost connections to your data drives. You have a couple of items to look at

Are the drives powered up ( and were they fully powered up when the system booted)

are all the cable connections secure?

If neither of the above help, you may have lost several disks, and need to to go into repair/recovery mode.

Dave

[Gurcharan Sahota_2]

I have checked the array, it's an MSA1000 and visually - all looks ok, no red lights.

Is there anything I can run to verify this?

[Gurcharan Sahota_2]

the output from the ioscan is as follows;

root@bddev:/dev/vg01> ioscan -fnC disk
Class I H/W Path Driver S/W State H/W Type Description
==========================================================================
disk 0 0/1/1/0.0.0 sdisk CLAIMED DEVICE HP 73.4GST373454LC
/dev/dsk/c2t0d0 /dev/rdsk/c2t0d0
disk 1 0/1/1/0.1.0 sdisk CLAIMED DEVICE HP 73.4GST373454LC
/dev/dsk/c2t1d0 /dev/rdsk/c2t1d0
disk 2 0/1/1/1.2.0 sdisk CLAIMED DEVICE HP DVD-ROM 30
5
/dev/dsk/c3t2d0 /dev/rdsk/c3t2d0
disk 6 0/4/1/0.100.0.0.0.0.0 sdisk CLAIMED DEVICE COMPAQ MS
A1000 VOLUME
/dev/dsk/c5t0d0 /dev/rdsk/c5t0d0
disk 7 0/4/1/0.100.0.0.0.0.1 sdisk CLAIMED DEVICE COMPAQ MS
A1000 VOLUME
/dev/dsk/c5t0d1 /dev/rdsk/c5t0d1
disk 8 0/4/1/0.100.0.0.0.3.4 sdisk CLAIMED DEVICE COMPAQ MS
A1000 VOLUME
/dev/dsk/c5t3d4 /dev/rdsk/c5t3d4

[Eric Antunes]

Hi,

Do you have PHKL_30607 and PHKL_32854 installed??

Regards,

Eric Antunes

[Steven E. Protter]

A most unfortuneate coincidence.

I would check with the administrator of the disks if this is on a SAN and make sure the ownership hasn't been toggled to a different world wide name.

The database parameter change surely did not cause this problem.

SEP

[Devesh Pant_1]

Please try to rescan the disks in. Also if you could provide the output of ioscan -fn so that the HBAs are also seen.

How does the lvmtab look. do a strings /etc/lvmtab and post the output for that too

thanks
Devesh

[Eknath]

Hi Gurcharan,

Compare you "ioscan -fnC disk" output with "strings /etc/lvmtab" output. check for the disk devices and confirm that all the devices mentioned in /etc/lvmtab are available in ioscan output. If something is missing that means some of your hard disks are not detected. Check for connections, power to storage etc.

Cheers!!!
eknath

[D Block 2]

Quorum implies that -> more than half of your volumes are missing.

can you back-out the Database change you made, and try another reboot..

verify you can see the phyical disks:
# ioscan -kfnC disk
# ioscan -fnC disk

also, maybe rebuild any special devices:
# insf -ev

is this a "shared" database with another server?

[Gurcharan Sahota_2]

Eric,

there is only patch PHKL_30607 installed on the system.

swlist -l patch | grep PHKL_30607
PHKL_30607.CORE-KRN 1.0 OS-Core.CORE-KRN applied
PHKL_30607.CORE2-KRN 1.0 OS-Core.CORE2-KRN applied

# PHKL_30607 1.0 VxVM,EMC,PM,vPar,slpq,KRS
,DUMP,LVM,VM,DaS,IF
# PHKL_30607.C-INC 1.0 ProgSupport.C-INC
applied
# PHKL_30607.CORE-KRN 1.0 OS-Core.CORE-KRN
applied
# PHKL_30607.CORE2-KRN 1.0 OS-Core.CORE2-KRN
applied
PHKL_30607.C-INC 1.0 ProgSupport.C-INC applied.

[Gurcharan Sahota_2]

Devesh,

I have attached the output as requested.

Could it be a patching issue? looking at a similar thread, we are missing some patches...

http://forums1.itrc.hp.com/service/forums/questionanswer.do?admit=716493758+1119004090278+28353475&threadId=881154

[scroll to the bottom], patch PHKL_30607 is already installed on the system.

[Gurcharan Sahota_2]

vgscan output

[Gurcharan Sahota_2]

ioscan bddev

[Gurcharan Sahota_2]

Tom,

have tried running a vgchange, results are as follows;

root@bddev:/> vgchange -a y /dev/vg01
vgchange: Volume group "/dev/vg01" does not exist in the "/etc/lvmtab" file.
root@bddev:/> vgchange -a y /dev/vg02
vgchange: Couldn't activate volume group "/dev/vg02":
Quorum not present, or some physical volume(s) are missing.

[Eric Antunes]

It seems your lvmtab is bad... Try this:

# mv /etc/lvmtab /etc/lvmtab.old
# vgscan -a

[shankar_3]

Hi Gurcharn,

As per output..

1.In the system vg01 and vg02 while activating its queery the PV belongs to VG.

2.Not able to find those so you have recevied above message.

possible way to find miised PV

1. Take the ouput of /etc/lvmtab

I have seen the your attachement in that vg01 entry not there and vg02 having one PV

It may possible that LVM header information get corrupted.

There could be other problem like PV's powerfailed, storage detached .

do following steps.

1.ioscan -fnC disk
2.mv /etc/lvmtab /etc/lvmtab.old
3.vgscan -v
4.activate all vgs
5.vgcfgbackup vgxx (taking all vg info backups)

Cheers,
Shankar

[Eric Antunes]

Hi,

Did "vgscan -a" worked?

Regards,

Eric Antunes

[Sunil Talreja]

I am having exact same problem, My machine model is same as yours.

Were you able to solve the problem.

[Sunil Talreja]

Since I didn't have any data, I just wiped everything and did a rebuilt of Volume group, Logical Volume and File system. Every thing seems to be working fine !!!