I came across the need to extend a partition on a Centos VM deployed in Azure which was not running LVM, but XFS. This seems to be a common way of how the templates are deployed in Azure.

I won’t go into detail on how you increase the disk space on the virtualization layer, I’ll leave that to you since it’s different for every platform and a fairly basic task.

This example is for increasing /dev/sda2. Let’s say you’ve increased your disk from 32GB to 64GB, however when issuing the below command to list your disks, you still see it at 32GB!

df -h

Resulting Output - 

Filesystem Size Used Avail Use% Mounted on
udev 1.5G 0 1.5G 0% /dev
tmpfs 291M 33M 259M 12% /run
/dev/sda2 32G 3.9G 32G 11% /
tmpfs 1.5G 4.0K 1.5G 1% /dev/shm
tmpfs 5.0M 0 5.0M 0% /run/lock
tmpfs 1.5G 0 1.5G 0% /sys/fs/cgroup
/dev/vda1 472M 171M 277M 39% /boot
tmpfs 291M 0 291M 0% /run/user/0

From here, you’ll want to run the below command to see that the OS is recognizing the increase space and just hasn’t allocated it – 

fdisk -l

Resulting Output - 

Disk /dev/vda: 64 GiB, 42949672960 bytes, 83886080 sectors

Now that we have confirmed the OS is seeing the extra space, lets assign it to the partition we need –

1) fdisk -l /dev/sda
2) p (This will print the existing partition table)
3) d (This will enter delete mode to remove the partition)
4) 2 (This is the partition number we want to delete)
5) p (This will print the partition table again to confirm)
6) n (Wizard to create new partition)
7) p (This selects type of partition, in our case P for primary)
8) 2 (Re use the partition number we had previously)
9) First sector - just press enter to accept the default
10) Last sector - same as above, this will ensure its assigned all available free space on that disk 
11) w - this will write the partition

After completing the above steps, you will be given a warning about the device or resource being busy, don’t worry this is completely fine. One more command before we reboot – 

partprobe

This will synchronize the partition to the kernel however fail due to requiring a reboot to take effect. Reboot now and execute the final step once back up –

xfs_growfs /

This will now assign all the free space available and we should be fine!