Resize2fs – Can’t Resize Partition

filesystemsresize2fsvirtual machinevmware

I recently resized the hard drive of a VM from 150 GB to 500 GB in VMWare ESXi. After doing this, I used Gparted to effectively resize the partition of this image. Now all I have to do is to resize the file system, since it still shows the old value (as you can see from the output of df -h):

Filesystem                     Size  Used Avail Use% Mounted on
/dev/mapper/owncloud--vg-root  157G   37G  112G  25% /
udev                           488M  4.0K  488M   1% /dev
tmpfs                          100M  240K  100M   1% /run
none                           5.0M     0  5.0M   0% /run/lock
none                           497M     0  497M   0% /run/shm
/dev/sda1                      236M   32M  192M  14% /boot

However, running sudo resize2fs /dev/mapper/owncloud--vg-root returns this:

resize2fs 1.42 (29-Nov-2011)
The filesystem is already 41608192 blocks long.  Nothing to do!

Since Gparted says that my partition is /dev/sda5, I also tried running sudo resize2fs /dev/sda5, but in this case I got this:

resize2fs 1.42 (29-Nov-2011)
resize2fs: Device or resource busy while trying to open /dev/sda5
Couldn't find valid filesystem superblock.

Finally, this is the output of pvs:

PV         VG          Fmt  Attr PSize   PFree
/dev/sda5  owncloud-vg lvm2 a-   499.76g 340.04g

fdisk -l /dev/sda shows the correct amount of space.

How can I resize the partition so that I can finally make the OS see 500 GB of hard drive?

Best Answer

If you only changed the partition size, you're not ready to resize the logical volume yet. Once the partition is the new size, you need to do a pvresize on the PV so the volume group sees the new space. After that you can use lvextend to expand the logical volume into the volume group's new space. You can pass -r to the lvextend command so that it automatically kicks off the resize2fs for you.

Personally, I would have just made a new partition and used vgextend on it since I've had mixed results with pvresize.

Related Solutions

Linux – Using resize2fs with file system offset

The offset option of mount does not get passed to mount directly, but to losetup which sets up a loop device which refers to the offsetted location of the underlaying block device. Mount then performs its operations on that loop device rather than the raw block device itself.

You can also use losetup to make resize2fs play which such file systems:

# losetup --offset=<offset> --find --show /dev/<device>
/dev/loop0
# resize2fs /dev/loop0 <newsize>
# losetup --detach /dev/loop0

(Example may not be complete in means of resize2fs operations)

losetup searchs for the first free loop device (in that example /dev/loop0) as --find was passed. --show outputs that loop device to STDOUT.

Resize root lvm partition

No, not safely. From what I have read, the VG must be available (vgchange --available y, but not mounted). I recommend that you boot from a liveCD or USB so that you have tools loaded into memory, but all of your filesystems not mounted.

Check the file system, we don't want to resize something in an inconsistent state

e2fsck -f /dev/s93079-vg/root

Resize the actual filesystem, a little smaller than the final target.

resize2fs /dev/s93079-vg/root 90G

Now, shrink the LVM to the desired size

lvreduce -L 100G /dev/s93079-vg/root

Finally, grow the the root filesystem to match the target LVM volume

resize2fs /dev/s93079-vg/root

While I browsed several guides, the most helpful blog I found on the topic was https://blog.shadypixel.com/how-to-shrink-an-lvm-volume-safely/

Best Answer

Related Solutions

Linux – Using resize2fs with file system offset

Resize root lvm partition

Related Question