Are you able to reproduce this issue (assuming you
can consistently) on current latest IceHouse RDO packages? You haven't
noted the versions you're using.

rpm -qa | grep icehouse

rdo-release-icehouse-4.noarch

Maybe you don't really have enough free space there? I don't have a
Cinder setup to do a sanity check, you might want to ensure if you have
your Cinder filter scheduler configured correctly.

I created another vg and gave it 30 gigs just in case:

[root@node1]/etc/cinder# (openstack_admin)] vgdisplay cinder-volumes

--- Volume group ---

VG Name cinder-volumes

System ID

Format lvm2

Metadata Areas 1

Metadata Sequence No 1

VG Access read/write

VG Status resizable

MAX LV 0

Cur LV 0

Open LV 0

Max PV 0

Cur PV 1

Act PV 1

VG Size 30.00 GiB

PE Size 4.00 MiB

Total PE 7679 Alloc PE / Size 0 / 0

Free PE / Size 7679 / 30.00 GiB
VG UUID huXkqD-3JIm-Fasr-Gkwc-EPVP-n15c-KIvBRc

> 2014-07-11 12:34:43.325 8665 ERROR cinder.scheduler.flows.create_volume
> [req-2acb85f1-5b7b-4b63-bf95-9037338cb52b 555e3e826c9f445c9975d0e1c6e00fc6
> ff6a2b534e984db58313ae194b2d908c - - -] Failed to schedule_create_volume:
> No valid host was found.
> 2014-07-11 12:35:16.253 8665 WARNING cinder.context [-] Arguments dropped
> when creating context: {'user': None, 'tenant': None, 'user_identity': u'-
> - - - -'}
>

--
/kashyap