Hi, so I managed to get Cinder to work with nfs, ceph and SMBFS creating
a multi-tier architecture. The permissions on the smbfs server are still
a small problem but when inherited and automatically changed, it works
very well.. also VERY fast.
On to the next issue.
I have glance images currently on ceph and thats ok but every time I
deploy a volume to the smb tier, it downloads the image the control node
(why?!?!?) and then creates the volume into smb or nfs .
Its very very very slow.
I tried to make a glance image-volume but I keep getting the following
error.
ERROR glance.common.wsgi AttributeError: 'Store' object has no attribute
'volume_id'
The default glance store is now cinder ( was rbd), all the cinder
options are enabled and yet its a no go.
Is there something else I need to change? The volume does exist on
cinder and if i change the name of it, i get the url was malformed so it
seems it finds the file but somehow it cannot add the volume id into the
metadata?
this is what i run:
openstack image create --disk-format raw --container-format bare
Linux-SMBFSTEST
which returns to me a uuid of: b97704ba-7b22-493d-a3bb-4ce597c9ddc2
and then I run
glance location-add --url cinder://e080103c-049e-4d91-9a3c-5322a505e5a4
b97704ba-7b22-493d-a3bb-4ce597c9ddc2
the cinder volume being volume-e080103c-049e-4d91-9a3c-5322a505e5a4 and
fully accessible with no permission restriction..
any help? anyone?
Thank you already
//florian
_______________________________________________
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
