increase vm disk size

[mndave668]

Hello,
I severely underestimated on of my vm's hard disk siize. Is there a way to increase its size from 100 gig to 500 gig without recreating the vm?

Thanks

 

[udo]

Hello,
I severely underestimated on of my vm's hard disk siize. Is there a way to increase its size from 100 gig to 500 gig without recreating the vm?

Thanks

Hi,
yes - it's depends on your disk-format/storage. If the storage is lvm-storage, you can simply expand the logical volume.
qcow2-files can you expand with "qemu-img resize filename +size".
If you use raw, you can also expand (append space on the raw-file), but be sure not to overwrite your data. Perhaps it's easier to create a new bigger disk and copy with dd the content from small-disk to big-disk.

Anyway you must after that resize the disk inside the vm - you can use an live-distro like partitionmagic for that.

Udo

 

[frater]

I had an OpenVZ server created with the Wordpress template.
When I had this problem I was amazed I could just resize the disk size using the Proxmox webIF.

Now I have the same problem and ran out of space (other VM)
I'm not able to do that again....

Why isn't it possible now?

 

[dietmar]

Why isn't it possible now?

What is the problem (exactly)?

 

[frater]

What is the problem (exactly)?

The harddisk is full on my OpenVZ system.
In the WebIF I set it to 140 GB, but the size never followed the size in the WebIF.
I did this before with another VM (but also based on the same template, "wordpress").
I believe it didn't even need a restart of the system.


The 2 machines have both a 150 GB harddisk and a 1 TB.
It is also running DRBD.



OpenVZ machine

root@zabbix:~# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/simfs             28G   28G     0 100% /
tmpfs                 3.9G     0  3.9G   0% /lib/init/rw
tmpfs                 3.9G     0  3.9G   0% /dev/shm
overflow              1.0M     0  1.0M   0% /tmp

Its Host:

proxmox-2:~# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/mapper/pve-root   35G  1.3G   32G   4% /
tmpfs                 3.9G     0  3.9G   0% /lib/init/rw
udev                   10M  744K  9.3M   8% /dev
tmpfs                 3.9G     0  3.9G   0% /dev/shm
/dev/mapper/pve-data   93G   93G     0 100% /var/lib/vz
/dev/sda1             504M   44M  435M  10% /boot

proxmox-2:~# blkid
/dev/sda1: UUID="13987169-1556-4900-91c2-f9278713c5b8" SEC_TYPE="ext2" TYPE="ext3"
/dev/sda2: UUID="jotiQ0-f4Bo-yKNd-M0LZ-eSGv-7SBl-vjS2Hg" TYPE="lvm2pv"
/dev/dm-0: TYPE="swap" UUID="52c04db1-a5c4-4ef5-8a96-655b4dda0f14"
/dev/dm-1: UUID="2ef9246f-8a3e-42a4-b114-936b663b14cc" TYPE="ext3"
/dev/dm-2: UUID="d75d2126-9670-4c11-a26f-e3186caf55c9" SEC_TYPE="ext2" TYPE="ext3"
/dev/sdb1: UUID="upE4dZ-ccwF-9GXc-ob9b-eaFY-FCzc-N244TH" TYPE="lvm2pv"
/dev/drbd0: UUID="upE4dZ-ccwF-9GXc-ob9b-eaFY-FCzc-N244TH" TYPE="lvm2pv"

 

[dietmar]

Well, the filesystem on the host is simple full!

/dev/mapper/pve-data   93G   93G     0 100% /var/lib/vz

 

[frater]

well, the filesystem on the host is simple full!

/dev/mapper/pve-data   93g   93g     0 100% /var/lib/vz

omg...

Of course it is....
I started to use proxmox for full virtualization and then learned about OpenVZ.
But these OpenVZ machines will go on the normal boot disk and not on the DRBD-volume, which I never realized because I was too busy with these machines themselves..

This also means I have no redundancy for OpenVZ machines :-(

Thanks, I will now quickly delete an obsolete VM on that host and get it going again....