Cannot Add alternate PV to a VG

[Tuck Lee]

I have use the DSF naming to create the Vg again;

root@ [/dev]
# mknod /dev/vg01/group c 64 0x010000

root@ [/dev]
# pvcreate -f /dev/rdisk/disk65
Physical volume "/dev/rdisk/disk65" has been successfully created.

root@ [/dev]
# vgcreate -s 8 -p 32 /dev/vg01 /dev/disk/disk65
Increased the number of physical extents per physical volume to 3236.
Volume group "/dev/vg01" has been successfully created.
Volume Group configuration for /dev/vg01 has been saved in /etc/lvmconf/vg01.conf

root@ [/dev]
# vgdisplay -v /dev/vg01
--- Volume groups ---
VG Name /dev/vg01
VG Write Access read/write
VG Status available
Max LV 255
Cur LV 0
Open LV 0
Max PV 32
Cur PV 1
Act PV 1
Max PE per PV 3236
VGDA 2
PE Size (Mbytes) 8
Total PE 3236
Alloc PE 0
Free PE 3236
Total PVG 0
Total Spare PVs 0
Total Spare PVs in use 0
VG Version 1.0
VG Max Size 809g
VG Max Extents 103552


--- Physical volumes ---
PV Name /dev/disk/disk65
PV Status available
Total PE 3236
Free PE 3236
Autoswitch On
Proactive Polling On



root@ [/dev]
# vgextend /dev/vg01 /dev/disk/disk95
Verification of unique LVM disk id on each disk in the volume group
/dev/vg01 failed.
Following are the sets of disks having identical LVM disk id
/dev/disk/disk65 /dev/disk/disk95

[Kapil Jha]

hmmmm can you please check from cstm or powermt if disk65 and disk95 are same disk or different disk.

Is it your DR server, does it have R2 devices present on the server.

I had the smthing similar issue and I could see the R2 devices on DR servers and hence the error.

RIght now you should confirm R2 device and one more thing, disk65 and disk95 cannot/should not be same disk.

BR,
Kapil+

[Tuck Lee]

disk65 and disk95 are the same LUN just from a different link path

/dev/rdisk/disk65 M(3) EMC SYMMETRIX 5670 15005C5000 26516160
/dev/rdisk/disk95 M(3) EMC SYMMETRIX 5670 15005C5000 26516160
/dev/rdsk/c13t0d3 M(3) EMC SYMMETRIX 5670 15005C5000 26516160
/dev/rdsk/c19t0d3 M(3) EMC SYMMETRIX 5670 15005C5000 26516160

[Kapil Jha]

smthing is wrong....there is no point having 2 dsf device for the same machine.

I would suggest to remove device file for disk65/95 and recreate them again.

BR,
Kapil+

[Kapil Jha]

>>>smthing is wrong....there is no point having 2 dsf device for the same machine.

It should be

smthing is wrong....there is no point having 2 dsf device for the same device.

BR,
Kapil+

[singh sanjeev]

can you see any scsimgr error in syslog.??

[chris huys_4]

Hi Tuck,

Deinstall powerpath and try again. In the past powerpath was not supported on HP-UX 11.31, so could be a reason for youre strange results..

Greetz,
Chris

[Tuck Lee]

Hi All,

What does all this means? if I have dual path in my server link to the DMX or clariion, in the DSF description i should see only one disk like "/dev/disk/disk65" only and no second disk link like in the lageacy lvm?

[R.O.]

Why don't you use the native multipathing with agile name convention instead of powerpath? I thought with native multipathing you could forget third party applications.

[Torsten.]

As long as the legacy device files are still enabled (disable them with rmsf, see man page), you still see 1 new device file and several legacy (one for each path).

[RC Park]

According to the documentation I have, there should no longer be any need for 3rd-party apps doing any multi-pathing. I've attached another document on this subject but here's a quote:

Multi-pathing is native to HP-UX 11i v3. It is built in to the mass storage subsystem and it is available to applications without any special configuration. Pre-11i v3 multi-pathing add-on products are no longer necessary on HP-UX 11i v3.
Native multi-pathing offers the following features:
â ¢
optimal distribution of I/O traffic across lunpaths to LUNs,
â ¢
dynamic discovery of lunpaths and LUNs,
â ¢
automatic monitoring of lunpaths,
â ¢
automatic lunpath failover and recovery,
â ¢
intelligent I/O retry algorithms to deal with failed lunpaths,
â ¢
lunpath authentication to avoid data corruption.

The attached document describes native multi-pathing in detail including examples. Hope this helps...

RCP

[Tuck Lee]

Hi All got some new information to share

# vgextend /dev/vg01 /dev/dsk/c19t8d1
Verification of unique LVM disk id on each disk in the volume group
/dev/vg01 failed.
Following are the sets of disks having identical LVM disk id
/dev/dsk/c15t8d1 /dev/dsk/c19t8d1

== > This is the known issue with EMC storage firmware.
== > Refer the below document for the reference.

--------------------------------------------
HP-UX 11.31 - LVM: vgextend Fails with "Verification of unique LVM disk id on each disk in the volume group failed"
ISSUE:
I created a volume group using persistent device files, but I cannot vgextend the alternate links into the volume group. The following error message is received:


# vgextend /dev/vg02 /dev/dsk/c8t0d5 /dev/dsk/c8t0d6 /dev/dsk/c8t0d7 /dev/dsk/c8t1d0
Verification of unique LVM disk id on each disk in the volume group /dev/vg02 failed.
Following are the sets of disks having identical LVM disk id
/dev/dsk/c6t0d5 /dev/dsk/c8t0d5
/dev/dsk/c6t0d6 /dev/dsk/c8t0d6
/dev/dsk/c6t0d7 /dev/dsk/c8t0d7
/dev/dsk/c6t1d0 /dev/dsk/c8t1d0



What is the source of this error? How do I correct it?

CONFIGURATION:
HP-UX 11.31
Storage array: EMC

SOLUTION:
Investigation of the error identified that the error message was due to a the version of firmware on the array.

Footnote 2 on page 7762 of the document titled EMC Support Matrix states the following:

"SPC-2 is required on Symmetrix FA director for HP-UX 11i v3

Check for:
SPC protocol revision = 4 .
If the revision is 2, the DMX SPC needs to have a firmware upgrade.

After upgrading the firmware (requires EMC assistance), the problem was resolved.

### END ###