[one-users] Problem with datastores after 4.4 to 4.6 upgrade

[Matthew Richardson]

I recently upgraded from 4.4 to 4.6 on a running cluster (using a ceph
datastore), and initially everything seemed to be fine, with existing
vms continuing to run.

However, if I know try to recreate an existing vm, or create a new one,
it gets stuck pending, and I see the following in sched.log:

Fri Jun  6 16:38:48 2014 [VM][D]: Pending/rescheduling VM and capacity
requirements:
      VM  CPU      Memory   System DS  Image DS
------------------------------------------------------------
      51  100     1048576           0  DS 101: 0
Fri Jun  6 16:38:48 2014 [HOST][D]: Discovered Hosts (enabled):
 0 1
Fri Jun  6 16:38:48 2014 [SCHED][D]: VM 51: Host 1 filtered out. Not
enough capacity.
Fri Jun  6 16:38:48 2014 [SCHED][I]: Scheduling Results:
Virtual Machine: 51

	PRI	ID - HOSTS
	------------------------
	-1	0

	PRI	ID - DATASTORES
	------------------------
	0	0

Fri Jun  6 16:38:48 2014 [SCHED][D]: VM 51: Local Datastore 0 in Host 0
filtered out. Not enough capacity.
Fri Jun  6 16:38:48 2014 [SCHED][I]: VM 51: No suitable System DS found
for Host: 0. Filtering out host.

onedatastore list shows:
ID NAME            SIZE AVAIL CLUSTER  IMAGES TYPE DS     TM
 0 system             - -     -            0 sys   -      ssh
 1 default         4.9G 62%   -            0 img   fs     shared
 2 files           4.9G 62%   -            0 fil   fs     ssh
101 ceph           10.9T 85%   -          19 img   ceph   ceph

vm template is:
NAME = "VM1"
CPU = "1"
VCPU = "1"
MEMORY = "3084"
DISK = [ IMAGE = "csx13126", DEV_PREFIX = "hd" ]

'oneimage show csx13126' gives:
ID             : 9
NAME           : csx13126
USER           : oneadmin
GROUP          : oneadmin
DATASTORE      : ceph
TYPE           : DATABLOCK
REGISTER TIME  : 10/09 17:20:37
PERSISTENT     : Yes
SOURCE         : one/one-9
FSTYPE         : raw
SIZE           : 48.8G
STATE          : used
RUNNING_VMS    : 1

I've no idea why it's trying (and failing) to use datastore 0, when the
disk image is on the ceph datastore (101).

Can anyone advise me on what is causing this, or any further steps I can
take to try and debug it?

Thanks in advance,

Matthew



-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: OpenPGP digital signature
URL: <http://lists.opennebula.org/pipermail/users-opennebula.org/attachments/20140606/65b68006/attachment.pgp>