vgimport behaves differently when giving persistent file names

[Kurz]

Hi,

I was just wondering, maybe you have come across this:

If I just do

vgimport -v -s -m /tmp/vg55.map,

then the volume group is based on the old non-persistent (legacy) device files, and I don't want that.

So I did:
# vgimport -v -m /tmp/vg55.map /dev/vg55 /dev/disk/disk88 /dev/disk/disk93
Creating "/etc/lvmtab_p".
Beginning the import process on Volume Group "/dev/vg55".
vgimport: Warning: Volume Group belongs to different CPU ID.
Can not determine if Volume Group is in use on another system. Continuing.
Logical volume "/dev/vg55/lvol1" has been successfully created
with lv number 1.
Logical volume "/dev/vg55/lvol2" has been successfully created
with lv number 2.
vgimport: Volume group "/dev/vg55" has been successfully created.
Warning: A backup of this volume group may not exist on this machine.
Please remember to take a backup using the vgcfgbackup command after activating the volume group.
#

My question is: Why does he not like the persistent device files in this case? He wasn't complaining when I imported with the -s option, but then the VG was based on the legacy paths.
# vgimport -v -m /tmp/vg55.map /dev/vg55 /dev/disk/disk88 /dev/disk/disk93
Creating "/etc/lvmtab_p".
Beginning the import process on Volume Group "/dev/vg55".
vgimport: Warning: Volume Group belongs to different CPU ID.
Can not determine if Volume Group is in use on another system. Continuing.
Logical volume "/dev/vg55/lvol1" has been successfully created
with lv number 1.
Logical volume "/dev/vg55/lvol2" has been successfully created
with lv number 2.
vgimport: Volume group "/dev/vg55" has been successfully created.
Warning: A backup of this volume group may not exist on this machine.
Please remember to take a backup using the vgcfgbackup command after activating the volume group.

[Carsten Krege · Accepted solution]

I'd say that this is not an effect of using new device files vs legacy device files.

In the map file you give with the vgimport -s -m

[Kurz]

You were right, -N is also what HP recommends:

"HP recommends converting volume groups with alternate links to use native multi-pathing by the use of persistent DSFs. The /usr/contrib/bin/vgdsf script, vgscan -N command, or vgimport -s -N commands perform this conversion. See the For More Information section for a whitepaper - LVM Migration from legacy to persistent naming model."

I had read the document but at that time i was only doing standard mirroring on 1 node, no ServiceGuard.

[Kurz]

Just to clarify what had happened:

I had exported the volume group with
# vgexport -v -p -s -m /tmp/vg55.map /dev/vg55

and then imported it on the 2nd node with

vgimport -s -m /tmp/vg55.map /dev/vg55

This gave me a volume group that looked like this:

# vgdisplay -v /dev/vg55
--- Volume groups ---
VG Name /dev/vg55
VG Write Access read/write
VG Status available
Max LV 255
Cur LV 2
Open LV 2
Max PV 16
Cur PV 2
Act PV 2
Max PE per PV 2559
VGDA 4
PE Size (Mbytes) 4
Total PE 5118
Alloc PE 5000
Free PE 118
Total PVG 0
Total Spare PVs 0
Total Spare PVs in use 0
VG Version 1.0
VG Max Size 163776m
VG Max Extents 40944

--- Logical volumes ---
LV Name /dev/vg55/lvol1
LV Status available/syncd
LV Size (Mbytes) 1000
Current LE 250
Allocated PE 500
Used PV 2

LV Name /dev/vg55/lvol2
LV Status available/syncd
LV Size (Mbytes) 9000
Current LE 2250
Allocated PE 4500
Used PV 2


--- Physical volumes ---
PV Name /dev/dsk/c1t6d7
PV Name /dev/dsk/c11t6d7 Alternate Link
PV Name /dev/dsk/c17t6d7 Alternate Link
PV Name /dev/dsk/c7t6d7 Alternate Link
PV Status available
Total PE 2559
Free PE 59
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c5t6d7
PV Name /dev/dsk/c3t6d7 Alternate Link
PV Name /dev/dsk/c15t6d7 Alternate Link
PV Name /dev/dsk/c13t6d7 Alternate Link
PV Status available
Total PE 2559
Free PE 59
Autoswitch On
Proactive Polling On

--> not what I wanted

Giving the persistent device files on the command lines instead made things better:

# vgchange -a n /dev/vg55
Volume group "/dev/vg55" has been successfully changed.
# vgexport /dev/vg55
vgexport: Volume group "/dev/vg55" has been successfully removed.
#

# vgimport -v -m /tmp/vg55.map /dev/vg55 /dev/disk/disk88 /dev/disk/disk93
Creating "/etc/lvmtab_p".
Beginning the import process on Volume Group "/dev/vg55".
vgimport: Warning: Volume Group belongs to different CPU ID.
Can not determine if Volume Group is in use on another system. Continuing.
Logical volume "/dev/vg55/lvol1" has been successfully created
with lv number 1.
Logical volume "/dev/vg55/lvol2" has been successfully created
with lv number 2.
vgimport: Volume group "/dev/vg55" has been successfully created.
Warning: A backup of this volume group may not exist on this machine.
Please remember to take a backup using the vgcfgbackup command after activating the volume group.
#
# vgchange -a y /dev/vg55
Activated volume group
Volume group "/dev/vg55" has been successfully changed.
# vgdisplay -v /dev/vg55
--- Volume groups ---
VG Name /dev/vg55
VG Write Access read/write
VG Status available
Max LV 255
Cur LV 2
Open LV 2
Max PV 16
Cur PV 2
Act PV 2
Max PE per PV 2559
VGDA 4
PE Size (Mbytes) 4
Total PE 5118
Alloc PE 5000
Free PE 118
Total PVG 0
Total Spare PVs 0
Total Spare PVs in use 0
VG Version 1.0
VG Max Size 163776m
VG Max Extents 40944

--- Logical volumes ---
LV Name /dev/vg55/lvol1
LV Status available/syncd
LV Size (Mbytes) 1000
Current LE 250
Allocated PE 500
Used PV 2

LV Name /dev/vg55/lvol2
LV Status available/syncd
LV Size (Mbytes) 9000
Current LE 2250
Allocated PE 4500
Used PV 2


--- Physical volumes ---
PV Name /dev/disk/disk88
PV Status available
Total PE 2559
Free PE 59
Autoswitch On
Proactive Polling On

PV Name /dev/disk/disk93
PV Status available
Total PE 2559
Free PE 59
Autoswitch On
Proactive Polling On

But it would probably have been best to use the -N option as you suggested.