[HessIBM]

Hello everyone, I just have a quick question about adding disk to my existing VIO/Client LPARs. Server is a 9113-550, with two VIO servers and two client LPARs. Currently, the server has only 4 physical harddrives. I am going to add 4 more physical drives to the server. I need to allocate more space to rootvg on the two client LPARs. How do I go about doing this? Here is an lsdev -virtual snapshot from 1 of the two VIO servers:

$ lsdev -virtual
name status description
ent2 Available Virtual I/O Ethernet Adapter (l-lan)
ent3 Available Virtual I/O Ethernet Adapter (l-lan)
vhost0 Available Virtual SCSI Server Adapter
vhost1 Available Virtual SCSI Server Adapter
vsa0 Available LPAR Virtual Serial Adapter
sappa1rootvg Available Virtual Target Device - Logical Volume
sappa2rootvg Available Virtual Target Device - Logical Volume
ent4 Available Shared Ethernet Adapter
ent5 Available Shared Ethernet Adapter

Any help anyone could provide would be most appreciated. If you need any more information from me to help provide a solution please let me know.

Thanks.

[cdelgadop]

Hi

First attach the new disks to the VIOS lpar. I'm assuming you're about to use the second 4-pack of internal disks so it's supposed that your p550 has a PCI SCSI adapter (could be a 5702) in the rear connected to the second backplane that allows using the below 4-pack and this slot is active in the VISO LPAR and configured. If so, then you need to go to the AIX enviroment in your VIOS by running:

oem_setup_env

and then run the following secuence of diag program:

diag->Task Selection (Diagnostics, Advanced Diagnostics, Service Aids, etc.)->Hot Plug Task-> SCSI and SCSI RAID Hot Plug Manager

In the "SCSI and SCSI RAID HOT PLUG MANAGER " you can list and attach the new disks to the server without actually reboot. Once configured, you can check this by running lsdev commands, you can exit form AIX and get into VIOS enviroment. Maybe you will need to run:

cfgdev -l vio0

in order for the vio0 device to configure the new disk. Again check this by running lsdev commands. Once new disks are config'd you can start running the procedure to create mapping and virtual devices on the VIOS server and assign to the different clients.

Hope this helps

[HessIBM]

Thanks very much for the information. I knew how to get the drives physically in the server using the hot swap. My question was more geared towards what to do after they were in. But once I got the drives in, and the server attempted to configure the scsi device they attached to..I received the following error:

Method error (etc/methods/cfgdisk -l hdiskX)
0514-086 Cannot perform the requested function because the specified device is not supported on this platform.

Upon further investigation, it turns out that this server has two 300GB 10K drives and two 73GB 10K drives. The drives that were purchased were 73GB 15K drives. I think this is a "no-no".

So it really turns out that we don't need these drives at all...we have plenty of space sitting on the server already.

I see from my VIO server that I have 250GB free. I need to increase the size of rootvg on my client LPARS, how can I do this using this free space that is shown on my VIO server but not shown on my client LPAR?

Sorry for the long post...I appreciate the help.

[cdelgadop]

Hi

It's your VIOS using Virtual SCSI Server Adapters mapped to Logical Volumes or storage pools or whole raw disks?.

If using Logical Volumes you can increase using the extendlv command at VIOS command level, then run

cfgdev -l vio0

and then in the AIX client you can run:

chvg -g <VGname>

you should then see the VG size increased.

Hope this helps

[HessIBM]

Here are the lsmap -all results from both of my VIO servers. I believe this indicates we are doing Vscsi adapters mapped to LVs:

VIO1:

$ lsmap -all
SVSA Physloc Client Partition
--------------- -------------------------------------------- -----------------
vhost0 U9113.550.10CED2F-V1-C110 0x00000002
VTD sappa1rootvg
LUN 0x8100000000000000
Backing device sappa1_rootvg
Physloc
SVSA Physloc Client Partition
--------------- -------------------------------------------- -----------------
vhost1 U9113.550.10CED2F-V1-C120 0x00000003
VTD sappa2rootvg
LUN 0x8100000000000000
Backing device sappa2_rootvg
Physloc

VIO2:

$ lsmap -all
SVSA Physloc Client Partition I
--------------- -------------------------------------------- ------------------
vhost0 U9113.550.10CED2F-V4-C210 0x00000002
VTD sappa1rootvg
LUN 0x8100000000000000
Backing device sappa1_rootvg
Physloc
SVSA Physloc Client Partition I
--------------- -------------------------------------------- ------------------
vhost1 U9113.550.10CED2F-V4-C220 0x00000003
VTD sappa2rootvg
LUN 0x8100000000000000
Backing device sappa2_rootvg
Physloc

Is this indicative of a VSCSI to LV mapping?

If so then all I need to do if I want to increase the rootvg on sappa1 is

extendlv sappa1_rootvg [SIZE] HDISK#? on VIO server

cfgdev -l vio0 on VIO server (is the last part of this command static or does it depend on what my vio server is named)

now do I perform the chvg -g [VG] command from the AIX prompt on VIO server or on the client LPAR?

[cdelgadop]

Hi

please post the output of:

lsvg

from VISO enviroment. Then

lsvg -lv for each VG

[HessIBM]

from VIO1

$ lsvg
rootvg
$ lsvg -lv rootvg
rootvg:
LV NAME TYPE LPs PPs PVs LV STATE MOUNT POINT
hd5 boot 1 2 2 closed/syncd N/A
hd6 paging 1 2 2 open/syncd N/A
paging00 paging 2 4 2 open/syncd N/A
hd8 jfs2log 1 2 2 open/syncd N/A
hd4 jfs2 1 2 2 open/syncd /
hd2 jfs2 4 8 2 open/syncd /usr
hd9var jfs2 2 4 2 open/syncd /var
hd3 jfs2 4 8 2 open/syncd /tmp
hd1 jfs2 20 40 2 open/syncd /home
hd10opt jfs2 1 2 2 open/syncd /opt
lg_dumplv sysdump 2 2 1 open/syncd N/A
sappa1_rootvg jfs 60 60 1 open/syncd N/A
sappa2_rootvg jfs 60 60 1 open/syncd N/A

from VIO2:

$ lsvg
rootvg
$ lsvg -lv rootvg
rootvg:
LV NAME TYPE LPs PPs PVs LV STATE MOUNT POINT
hd5 boot 1 2 2 closed/syncd N/A
hd6 paging 1 2 2 open/syncd N/A
paging00 paging 2 4 2 open/syncd N/A
hd8 jfs2log 1 2 2 open/syncd N/A
hd4 jfs2 1 2 2 open/syncd /
hd2 jfs2 4 8 2 open/syncd /usr
hd9var jfs2 2 4 2 open/syncd /var
hd3 jfs2 4 8 2 open/syncd /tmp
hd1 jfs2 20 40 2 open/syncd /home
hd10opt jfs2 1 2 2 open/syncd /opt
lg_dumplv sysdump 2 2 1 open/syncd N/A
sappa1_rootvg jfs 60 60 1 open/syncd N/A
sappa2_rootvg jfs 60 60 1 open/syncd N/A
$

[cdelgadop]

Hi

As you can see there are two logical volumes LVs (sappa1_rootvg and sappa2_rootvg) and those are mapped to the Virtual SCSI server adapters (vhost0 and vhost1) which in turn are the Virtual SCSI client adapters for each client. So if you issue the ommand:

extendlv <LVNAME> <Size in Logical Partitions> <Disk_where you want to grow>

Once this operation is OK. You can go to the client LPAR and from root issue:

getconf DISK_SIZE /dev/hdiskXX

Where hdiskXX is the virtual disk from the VIOS Server. This command below will show the size of the disk. The issue:

cfgmgr

Then

lsvg <VGNAME>

where VGNAME is the Volume group whose disks you want to increase.

chvg -g <VGNAME>

Check that more free space is available.

In fact the disk size should be larger.

Hope this helps

Question: Do you have backups of mksysb and data right ???

[HessIBM]

Thank you very much. I have recent mksysbs of the LPARs. I also have backups of the VIO servers from when I upgraded the IOSlevel for daylight savings time changes. Thanks again for all the help, i will be sure to let u know how I make out

[HessIBM]

the man page for chvg -g says that resizing of rootvg is not supported, this is exactly what I am trying to do????