Glance needs some additional permissions including write access to the pool you want to add images to. See the docs at:
http://ceph.com/docs/master/rbd/rbd-openstack/
Cheers,
Erik
On Apr 6, 2015 7:21 AM, <florian.rommel@xxxxxxxxxxxxxxx> wrote:
Hi, first off: long time reader, first time poster :)..
I have a 4 node ceph cluster (~12TB in total) and an openstack cloud
(juno) running.
Everything we have is Suse based and ceph 0.80.8
Now, the cluster works fine.. :
cluster 54636e1e-aeb2-47a3-8cc6-684685264b63
health HEALTH_OK
monmap e1: 3 mons at
{ceph01=10.70.0.100:6789/0,ceph03=10.70.0.102:6789/0,ceph04=10.70.0.103:6789/0},
election epoch 6, quorum 0,1,2 ceph01,ceph03,ceph04
osdmap e40: 7 osds: 7 up, 7 in
pgmap v78: 447 pgs, 5 pools, 0 bytes data, 0 objects
254 MB used, 12986 GB / 12986 GB avail
447 active+clean
I also have pools for images and volumes ready:
ceph04:~ # ceph osd lspools
0 data,1 metadata,2 rbd,3 volumes,4 images,
and i have the keyrings and permissions done:
client.admin
key: XXXXXXXXXXXXX
caps: [mds] allow
caps: [mon] allow *
caps: [osd] allow *
client.bootstrap-mds
key: XXXXXXXXXXXXX
caps: [mon] allow profile bootstrap-mds
client.bootstrap-osd
key:XXXXXXXXXXXXX
caps: [mon] allow profile bootstrap-osd
client.glance
key: XXXXXXXXXXXXX
caps: [mon] allow r
caps: [osd] allow class-read object_prefix rbd_children, allow rwx \
pool=images
client.volumes
key: XXXXXXXXXXXXX
caps: [mon] allow r
caps: [osd] allow class-read object_prefix rbd_children, allow rwx \
pool=volumes
I have copied the files to the openstack glance server and in the
ceph.conf, added the keyring sections.
mon_initial_members = ceph01, ceph03, ceph04
mon_host = 10.70.0.100,10.70.0.102,10.70.0.103
auth_cluster_required = cephx
auth_service_required = cephx
auth_client_required = cephx
filestore_xattr_use_omap = true
[client.glance]
keyring=/etc/glance/ceph.client.glance.keyring
The glance user has permissions to read the files.
Now,
when i execute this command:
glance image-create --name CIRROS --is-public true --disk-format qcow2
--container-format bare --file cirros-0.3.3-x86_64-disk.img
i get as a response:
<h1>500 Internal Server Error</h1>
Failed to upload image 3fc9fe83-cc52-4481-b95c-2b5724c1d971<br /><br />
and in /var/log/glance/api.log I get this:
2015-04-06 14:15:49.097 15203 TRACE glance.api.v1.upload_utils
features=rbd.RBD_FEATURE_LAYERING)
2015-04-06 14:15:49.097 15203 TRACE glance.api.v1.upload_utils File
"/usr/lib64/python2.6/site-packages/rbd.py", line 219, in create
2015-04-06 14:15:49.097 15203 TRACE glance.api.v1.upload_utils raise
make_ex(ret, 'error creating image')
2015-04-06 14:15:49.097 15203 TRACE glance.api.v1.upload_utils
PermissionError: error creating image
2015-04-06 14:15:49.097 15203 TRACE glance.api.v1.upload_utils
I am a bit stumped... omn the ceph cluster I see nothing in the logs. its
almost as if it wont even leave from the glance server..
Any ideas here? I would really appreciate it..
Thanks already,
//f
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
