Expanding volume size for LVM volumes

If you expand a block storage volume that is attached to a virtual machine by using the Default attach block disk add-on that is greater than or equal to 2 TB that was previously formatted to Logical Volume Manager (LVM), you must complete additional steps.

About this task

If you are using Red Hat Enterprise Linux® 6.x and if you expand a block storage volume to greater than or equal to 2 TB, you must resize the logical volume that is mounted on the virtual machines manually. During this manual expansion, you must detach the volume or restart the virtual instance. Therefore, you might want to complete this task during a maintenance window.

If you expand the logical volume to 2 TB or more on Red Hat Enterprise Linux 6.x, during a Refresh action or during a new deployment that uses an add-on, the following error message is produced: The block storage device size expansion is not reflected in the mounted logical volume. Automatic logical volume size refresh is not supported on RHEL 6.x if the size is expanded to 2 TB or more. To refresh the logical volume size, manually rescan the device, create a new partition, restart or the detach volume, and then add the partition to a physical volume, volume group, logical volume and refresh the mount.

After you expand the size of the external volume and then rediscover the size change from the Hardware > Storage Resources page in the Cloud Pak System Software console, log in to the virtual machine by using SSH and complete the following steps.

Note: You might have to also change the size in PowerVC.

Procedure

Identify the mount point that is associated with the block storage from the pattern instance block storage table.
Run df -Ph to get the logical volume file system name that corresponds to the mount point. The file system name is in the following format: /dev/mapper/<volume_group_name>-<logical_volume_name>.
For example, if your mount point is /mount1, when you run df -Ph you must see something similar to the following output:
```
Filesystem                   Size  Used  Avail Use%  Mounted on
/dev/mapper/datavg0-LV21328  3.0G  1.6M  2.8G  1%    /mount1
```
Get the volume group name and logical volume name from the output of the command that you ran in the previous step.
In the example in the previous step, the volume group name is datavg0 and logical volume name is LV21328.
Run the pvs command to get the physical device that is associated with the volume group. The pvs command returns the physical device with the partition number.
When you run the pvs command, you see output similar to the following example:
```
   PV         VG      Fmt   Attr   PSize       PFree 
   /dev/sda2  rhel    lvm2  a--    11.75g      0
   /dev/sdb1  datavg0 lvm2  a--    1020.00m    0      
   /dev/sdb2  datavg0 lvm2  a--    1020.00m    0
```
If your volume group is datavg0 then your physical device is /dev/sdb
Rescan the device to get the new size: echo 1 > /sys/block/<PHYSICAL_DEVICE>/device/rescan.
For example, echo 1 > /sys/block/sdb/device/rescan.
Run the following command to fix any alignment errors: parted /dev/sdb print. When you are prompted to fix or ignore errors, type Fix.

Run the following command to get the start and end values for expanded disk segment: parted -s /dev/<PHYSICAL_DEVICE> u MB p free. The expanded disk segment is the last entry in the output.

The preceding command returns output similar to the following example:

parted -s /dev/sdb u MB p free
Model: IBM 2145 (scsi)
Disk /dev/sdb: 3221MB
Sector size (logical/physical): 512B/512B
Partition Table: gpt
Disk Flags:
Number  Start     End       Size      File system  Name          Flags          
        0.02MB   1.05MB   1.03MB                   Free Space 
1       1.05MB   1073MB   1072MB                   P1            lvm      
2       1073MB   2147MB   1075MB                   P2            lvm        
        2147MB   3221MB   1074MB                   Free Space

The start value is 2147 MB and the end value is 3221 MB.

Run the following command to partition: parted -s -a optimal /dev/<PHYSICAL_DEVICE> mkpart <Partition_Name> <start> <end>. The start and end value must be the same as the values that you retrieved in the previous step (including the trailing MB string).
The following example creates a new partition with the start and end value that was obtained from previous step:parted -s -a optimal /dev/sdb mkpart P3 2147MB 3221MB

Run the following command to set the lvm flags on the new partition: parted /dev/<PHYSICAL_DEVICE> set <partition_number> lvm on.

For example, to get the new partition number run parted /dev/sdb print. The output is:

Number  Start     End       Size      File system   Name         Flags          
        0.02MB    1.05MB    1.03MB                  Free Space 
1       1.05MB    1073MB    1072MB                  P1           lvm
2       1073MB    2147MB    1075MB                  P2           lvm 
3       2147MB    3221MB    1074MB                  P3

To set partition 3 to lvm, run: parted /dev/sdb set 3 lvm on.

Detach and reattach the volume from pattern instance page or restart the virtual machine so that the kernel can identify the new partition.
After the volume is reattached or restarted, run ls /dev/sd*. You should see the new partition in the command output.
For example, the following command displays the new partition:
```
/dev/sdb3 ls /dev/sdb* 
/dev/sdb /dev/sdb1 /dev/sdb2 /dev/sdb3
```
Run the following command to create a new physical volume from the new partition: pvcreate /dev/<PHYSICAL_DEVICE><PARTITION_NUMBER>
For example, to add /dev/sdb3 run: pvcreate /dev/sdb3.
Run the following command to add the new physical volume to the volume group that is associated with the block storage: vgextend <volume_group_name> /dev/<PHYSICAL_DEVICE><PARTITION_NUMBER>. To get the volume group that is associated with the block storage, see step 2 and 3.
For example, if the block storage is associated with volume group datavg0, to extend volume group datavg0 with /dev/sdb3 run: vgextend datavg0 /dev/sdb3.
Run the following command to extend the logical volume that is associated with the block storage: lvextend -l +100%FREE /dev/<volume_group_name>/<logical_volume_name>. To get the logical volume that is associated with the block storage, see step 2 and 3.
For example, if the block storage is associated with volume group datavg0 and logical volume LV21328, to extend the logical volume run lvextend -l +100%FREE /dev/datavg0/LV21328.
Run the following command to refresh the size of the file system: resize2fs /dev/<volume_group_name>/<logical_volume_name>.
For example, if the block storage is associated with volume group datavg0 and logical volume LV21328, to refresh the file system run: resize2fs /dev/datavg0/LV21328.
To confirm that the file system is resized, run df -Ph and you should see the new size.