Cinder is creating images of about TB instead of GB when using ceph

[gadLinux]

Hi,

I just created an image of about 8GB of persistent storage. But cinder created this:

ubuntu@web-server:~$ df
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/vda1 2113757104 766568 2027062560 1% /
none 4 0 4 0% /sys/fs/cgroup
udev 504016 12 504004 1% /dev
tmpfs 101796 332 101464 1% /run
none 5120 0 5120 0% /run/lock
none 508976 0 508976 0% /run/shm
none 102400 0 102400 0% /run/user

it took a while to create it.

Fdisk shows:

WARNING: The size of this disk is 8.8 TB (8796093022208 bytes).
DOS partition table format can not be used on drives for volumes
larger than (2199023255040 bytes) for 512-byte sectors. Use parted(1) and GUID
partition table format (GPT).

Command (m for help): p

Disk /dev/vda: 8796.1 GB, 8796093022208 bytes
4 heads, 32 sectors/track, 134217728 cylinders, total 17179869184 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00001999

   Device Boot Start End Blocks Id System
/dev/vda1 * 2048 4294967295 2147482624 83 Linux

This is a real problem since it can leave cluster without space fast if goes unnoticed.

I'm using openstack icehouse in ubuntu 14.04 LTS.

Linux red-compute 3.13.0-24-generic #47-Ubuntu SMP Fri May 2 23:30:00 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux

[gadLinux (gad-aguilardelgado)]

Note:

The flavor seems to be configured ok:

normal.server 1 1024MB ***8GB*** 0GB 1024MB b9808074-0218-4277-a667-a4caa8957ffb

[gadLinux (gad-aguilardelgado)]

A workaround is to boot from a volume and set there the volume size. This is right done.

[David Medberry (med)]

Confirming bug. Seeing this in Icehouse with dumpling.