VG extend USING EMC STORAGE

[shabash]

I have 2 vggroup assigned from EMC storage in cluste envirnment.
One VG is showing as full so i need to add space from other vg group which have space available.
we need to vgreduce first and then vgextend for low volum vg.

kinldy some one share the procedure to or commands to follow or check before doing this activity.

[Ganesan R]

Hi Shabash,

Find out if any disk is free on the second volume group. You can use vgdisplay -v vg2 or pvdisplay -v disk to find if any PE's are used by any logical volume.

if all the PE's are free, use vgreduce command to remove the disk from the second volume group. Syntax would be,

#vgreduce vg2 disk

Once you removed the disk from vg02 you can add it to first volume group using vgextend. Syntax would be,

#vgextend vg1 disk

Now you can use the disk space for vg1

[Ganesan R]

Hi Shabash,

Could you provide # vgdisplay -v vgname output from both volume groups ?

[Lijeesh N G_1]

Hi Shabash,


1) Make sure the PV is not used for any LV.

#pvdisplay


# vgdisplay -v vg02
--- Volume groups ---
VG Name /dev/vg02
VG Write Access read/write
VG Status available
Max LV 255
Cur LV 0
Open LV 0
Max PV 16
Cur PV 2
Act PV 2
Max PE per PV 4375
VGDA 4
PE Size (Mbytes) 32
Total PE 8748
Alloc PE 0
Free PE 8748
Total PVG 0
Total Spare PVs 0
Total Spare PVs in use 0


--- Physical volumes ---
PV Name /dev/dsk/c2t0d0
PV Status available
Total PE 4374
Free PE 4374
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c3t2d0 ====> Take this for eg.
PV Status available
Total PE 4374
Free PE 4374
Autoswitch On
Proactive Polling On

# pvdisplay /dev/dsk/c3t2d0
--- Physical volumes ---
PV Name /dev/dsk/c3t2d0
VG Name /dev/vg02
PV Status available
Allocatable yes
VGDA 2
Cur LV 0 <<<==========
PE Size (Mbytes) 32
Total PE 4374
Free PE 4374
Allocated PE 0 <<<==========
Stale PE 0
IO Timeout (Seconds) default
Autoswitch On
Proactive Polling On





2)Remove the PV from VG; reduce the VG

#vgreduce

# vgreduce /dev/vg02 /dev/dsk/c3t2d0
Volume group "/dev/vg02" has been successfully reduced.
Volume Group configuration for /dev/vg02 has been saved in /etc/lvmconf/vg02.conf

3)Verify with vgdisplay

# vgdisplay -v /dev/vg02
--- Volume groups ---
VG Name /dev/vg02
VG Write Access read/write
VG Status available
Max LV 255
Cur LV 0
Open LV 0
Max PV 16
Cur PV 1
Act PV 1
Max PE per PV 4375
VGDA 2
PE Size (Mbytes) 32
Total PE 4374
Alloc PE 0
Free PE 4374
Total PVG 0
Total Spare PVs 0
Total Spare PVs in use 0


--- Physical volumes ---
PV Name /dev/dsk/c2t0d0
PV Status available
Total PE 4374
Free PE 4374
Autoswitch On
Proactive Polling On <<<====/dev/dsk/c3t2d0 is removed



4) Add PV to to the other VG

#vgextend

# vgextend /dev/vg01 /dev/dsk/c3t2d0
Volume group "/dev/vg01" has been successfully extended.
Volume Group configuration for /dev/vg02 has been saved in /etc/lvmconf/vg01.conf

5) Verify with vgdisplay

# vgdisplay -v 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 16
Cur PV 2
Act PV 2
Max PE per PV 4375
VGDA 4
PE Size (Mbytes) 32
Total PE 8748
Alloc PE 0
Free PE 8748
Total PVG 0
Total Spare PVs 0
Total Spare PVs in use 0


--- Physical volumes ---
PV Name /dev/dsk/c3t0d0
PV Status available
Total PE 4374
Free PE 4374
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c3t2d0 <<<<======Added /dev/dsk/c3t2d0
PV Status available
Total PE 4374
Free PE 4374
Autoswitch On
Proactive Polling On


6)If the VG is in Cluster, you have to export from present node and import to another node.

Export configuration in present Node

Node1:
-----
#vgchange -a n vg02
#vgexport -p -s -m /tmp/vg02.map

Copy map file to another cluster node
#scp /tmp/vg02.map Node2:/tmp/vg02.map

Import in another cluster node
#vgimport -p -s -m /tmp/vg02.map


Reagrds,
LIJEESH N G

[Ganesan R]

Hi Shabash,

First of all, you cannot free up any space from both the volumes. Because both volume group has single LV which is distributed across all the physical volumes. So unless you remove the logical volume or recreate the filesystem with small size, you cannot reclaim the space from any of the physical disk.

[Ganesan R]

Hi Again,

I would suggest you to order additional luns from storage team to extend the VG if possible.

[Lijeesh N G_1]

Hi,

I could see there are 4 PV's for bith VG's vgdb and vgbackup. Again one LV for each VG's with striped across all 4 available PV's. So there is no free PV in both the VG's. You cannot reduce any of the VG.

Regards,
LIJEESH N G

[Lijeesh N G_1]

Hi,

Either you have to remove the LV and recreate it with 3 PV's or assign a new LUN and use it for this activity.

Regards,
LIJEESH N G

[Lijeesh N G_1]

Hi,

Some texts are missing in above reply!!!!

Either you can remove the LV and recreate with 3 PV's or assign a new LUN and use it for this activity.

Regards,
LIJEESH N G

[shabash]

Ok i will proceed with removing LV.

Kindly tell me step by step command procedure to proceed.

Also tell me step by step cluster configuration package halt and vg export and import command with each step.

Any one
???????

[Rita C Workman]

Shabash,

Believe it or not, I have your interests in mind. The first thing you need to do is either attend a class and open a book and read. Talk to your boss and see if they will pay for some classes to help you learn to do the job of an HPUX Admin. Because it is unreasonable to expect you to do it otherwise.
LVM tasks are basic to the duties of any HPUX Admin. Clusters involve a bit more complexity, but for your situation the first LVM steps are the same. After you re-allocate your disk between vg's, you just create a fresh mapfile, send it to the other nodes in the cluster and re-import it so they are up to date.

But, if you want to be a good sys admin you need to start by learning and loving learning. Build on every new discovery and task you master.
This site was intended to help when folks run into 'issues' on something. It was not intended to be, although lately it is exactly what it is being dragged into being, a site to teach the basics for everyone who thinks they want to be an HPUX Admin. It's been done so much already - just type your need in the Search utility.

Old saying ".. give a man a fish and you feed him for a day, teach him to fish and you feed him for life". The book is your line, cast it into the water of your mind, it will serve you best in your efforts to learn.

And we will be here if you run into an issue and need help.

Kindest regards,
Rita

[James R. Ferguson]

Shabash:

For the sake of everyone, PLEASE post LARGE amounts of text as an ATTACHMENT (as a simple ASCII file)!

...JRF...

[Jim Turner]

Shabash,

I'm compelled to echo Rita's and Jim's sound wisdom. I'm also trying to be as kind as possible.

Forums participation presumes a baseline literacy in HP-UX sysadmin tasks including LVM. Quite franky, anyone without basic sysadmin training -- and more importantly, experience -- shouldn't possess the root password and certainly shouldn't have admin access to SAN storage where one wrong keystroke could wipe production data.

If a seasoned, senior sysadmin is not available to mentor you in your workplace, please do as Rita has advised and seek training.

Respectfully,
Jim

[shabash]

Thanks for your advices.

My purpose is to get maximum information from you guys so that no step will be missed out from my side.

i have create list of task to perform plz check and confirm for anything missing.

Total nodes in cluster =3=rc03ds,rc03ss.rc03cs1
OS= hp-ux v1
Node on which vg extend=rc03ds
Package on rc03ds=db and base-rc03ds
VG to extend=/dev/vgdb=total disks=4
1 path from EMC=500GB
4 path=2TB approx
Requirement=500GB to VGdb
Strings /etc/lvmtab
/dev/vgdb
/dev/dsk/c5t1d0
/dev/dsk/c10t1d1
/dev/dsk/c5t1d3
/dev/dsk/c10t1d4
/dev/dsk/c10t1d0
/dev/dsk/c5t1d1
/dev/dsk/c10t1d3
/dev/dsk/c5t1d4

VG to remove=/dev/VGbackup=total disks=4
Strings /etc/lvmtab
/dev/vgbackup
/dev/dsk/c12t0d0
/dev/dsk/c13t0d1
/dev/dsk/c12t0d2
/dev/dsk/c13t0d3
/dev/dsk/c13t0d0
/dev/dsk/c12t0d1
/dev/dsk/c13t0d2
/dev/dsk/c12t0d3

VGDB After extend =total disk will be 5=rest 3 disk remain as unused or add with VGbackup.

On node rc03ds:
Un mount the /d/backup filesystem

# umount /d/backup

De-activate the VG.
# vgchange -a n vgbackup

Export the VG.
# vgexport vgbackup

pvcreate on the disks with -f option for extending to vgdb

# pvcreate -f /dev/dsk/c12t0d0

Extend the VG vgdb with the disks of vgbackup.
Confirm the disk path you have written below
# vgextend vgdb /dev/dsk/c12t0d0 /dev/dsk/c13t0d0 <<< Alt path for /dev/dsk/c12t0d0

Size increase of vgdb logical volume multilple of what??
As required size is 500GB so how much with mb
# lvextend -L "new value in MB" /dev/vgdb/lvol1
Here downtime means using single user mode at system startup or on running system
As there is no online JFS installed in the system so i need the downtime for the application using the "/d/db" file system.

Unmount the /d/db file system
# umount /d/db

Run the extendfs on the filesystem with raw logical volume.

# extendfs /dev/vgdb/rlvol1

Mount the file system and check with bdf command for successful extension.
# mount /dev/vgdb/lvol1 /d/db
# bdf

# rcp /etc/lvmconf/vgdb.map rc03cs1:/etc/lvmconf
# rcp /etc/lvmconf/vg07.map rc03ss:/etc/lvmconf


On the other node(s) where the VG is not active:

Record the group minor number for the VG to which i added the disk.

Example:

# ll /dev/vg*/group


crw-r----- 1 root root 64 0x000000 Sep 19 1996 /dev/vgdb/group
crw-rw-rw- 1 root sys 64 0x020000 Jun 4 12:39 /dev/vg00/group
^^^^^^^^
record above minor number

vgexport /dev/vgdb

# mkdir /dev/vgdb/
# mknod /dev/vgdb/group c 64 0x020000

Reimport the VG using the new map file.

# vgimport -vs -m /etc/lvmconf/vgdb.map /dev/vgdb