Alternate Link

[warghade]

When i am trying to configure the alternate link to VG vgtr,

# vgextend vgtr /dev/dsk/c2t1d0

getting the following massage:

vgextend: Physical Volume "/dev/dsk/c2t1d0" contains no LVM information.

VG vgtr is currently on /dev/dsk/c1t1d0 and
/dev/dsk/c2t1d0 & /dev/dsk/c2t1d0 represents the same disk.

[Manju Kampli]

Double check the hardware address of both disks and make sure that both are pointing to the same physical disks.

1. Perform diskinfo on both disks and look for size
2. Physically check that, the disk is being shared across the both channels.

[Andreas Voss]

Hi,

what is the output from:
pvdisplay /dev/dsk/c1t1d0
and
pvdisplay /dev/dsk/c2t1d0
Could you see the devices by
ioscan -fnC disk ?
Please post the vgdisplay -v vgtr to the forum to see more detail.

Regards

Andrew

[warghade]

Currnetly VG vgtr is on /dev/dsk/c1t0d0.
# pvdisplay /dev/dsk/c1t0d0
--- Physical volumes ---
PV Name /dev/dsk/c1t0d0
VG Name /dev/vgtr
PV Status available
Allocatable yes
VGDA 2
Cur LV 13
PE Size (Mbytes) 4
Total PE 12615
Free PE 0
Allocated PE 12615
Stale PE 0
IO Timeout (Seconds) default

# pvdisplay /dev/dsk/c2t1d0
pvdisplay: Couldn't find the volume group to which
physical volume "/dev/dsk/c2t1d0" belongs.
pvdisplay: Cannot display physical volume "/dev/dsk/c2t1d0".

# ioscan -fnCdisk
Class I H/W Path Driver S/W State H/W Type Description
=====================================================================
disk 0 8/4.5.0 sdisk CLAIMED DEVICE SEAGATE ST34572WC
/dev/dsk/c0t5d0 /dev/rdsk/c0t5d0
disk 1 8/4.8.0 sdisk CLAIMED DEVICE SEAGATE ST34572WC
/dev/dsk/c0t8d0 /dev/rdsk/c0t8d0
disk 2 8/8.0.0 sdisk CLAIMED DEVICE DGC C2300WDR5
/dev/dsk/c1t0d0 /dev/rdsk/c1t0d0
disk 3 8/8.0.1 sdisk CLAIMED DEVICE DGC C2300WDR5
/dev/dsk/c1t0d1 /dev/rdsk/c1t0d1
disk 4 8/12.1.0 sdisk CLAIMED DEVICE DGC C2300WDR5
/dev/dsk/c2t1d0 /dev/rdsk/c2t1d0
disk 6 8/16/5.2.0 sdisk CLAIMED DEVICE TOSHIBA CD-ROM XM-5701T
A

[melvyn burnard]

from the information you have provided, I believe you have a NIKE array connected to the system.
If this is so, I also believe that you do not have auto trespass set on on the array, as I believe the array will not allow you to present a pv with the scsi address different on the two array controllers, i.e. c1t0d0 should be c2t0d0 and c1t1d0 should be seen as c2t2d0.
I think I am right in this assumption, but it is a long timew since I played with a NIKE array. If this is the case, then the disk at c1t0d0 is NOT the same as the disk at c2t1d0.
check your array for it's configuration settings.

[CHRIS_ANORUO]

Hi,

Do a vgchange -a y /dev/vgtr. Use the command strings /etc/lvmtab check the output for /dev/vgtr. Copy /etc/lvmtab to /etc/lvmtab.2407 and then use vgscan to scan all physical volumes in order to allow for recovery of the lvm configuration file (/etc/lvmtab).

[Tim Malnati]

Just another quick item. LVM commands want the full absolute path to all devices on the command line. Your command should be # vgextend /dev/vgtr /dev/dsk/c2t1d0

[warghade]

Yet no improvement...

Thanks to all of you.

[Emmanuel Eyer]

Okay, you have a Nike.

As pointed out, you MUST have autotrespass feature enabled. From your ioscan it seems it is not the case (LUN 0 is always "seen" on both controllers (SPs), but not accessible). To enable autotrespass, set your system option package to 2 (can be done from Grid Manager, in FE mode I believe).

Once autotrespass is enabled, you should have no problem to do your vgextend.

One important point: be sure to set the primary link thru the controller (SP) which is the DEFAULT owner of the LUN (by default the SP which created the LUN). Otherwise you will have performance issues.

Hope it helps, Emmanuel

[R D SUNDAR RAJAN]

For a disk which has a primary and an alternate link only the controller part of the the device file will change that is cXtYdZ Here Y and Z will remaim the same.
From your pvdisplay it says the device file as /dev/dsk/c1t0d0 so the alternate link may be /dev/dsk/c2t0d0 so try to extend the vg ti this device file.

[Bill McNAMARA_1]

Have a look at this response for all the info you need relating to configuration of a Nike array.

http://forums.itrc.hp.com/cm/QuestionAnswer/1,1150,0x696e7e990647d4118fee0090279cd0f9,00.html

Later,
Bill