If you're trying to do what I think you are, which is to create a bootable cinder volume (in rbd) using a glance image (in rbd), you can enable the following in glance-api.conf:
show_image_direct_url = True Enabling this will allow cinder to 'see' that the glance image is also in rbd, and it will essentially make a fast copy-on-write clone of the glance image into your cinder volume (effectively it's just an rbd clone from one pool, images, into another, volumes). - Darren On 27 March 2014 10:46, Thomas Bernard <t...@atolcd.com> wrote: > Yes i have a 10G network between all my node (controller, compute & > storage). > I precise that i create new instance with "boot from image (create new > volume)". > > -- > Le 27/03/2014 11:11, Ken Peng a écrit : > > 于 2014-3-27 17:21, Thomas Bernard 写道: >> >>> I use a Havana infrastructure with CEPH storage (cinder & rbd) and when >>> i create new big volume this is very slow. >>> Openstack provision all the space on CEPH cluster this is very very >>> (very) slow. >>> >> >> Do you have the enough bandwidth from compute nodes to the storage node? >> For example, a 10G+bps connection link. >> >> _______________________________________________ >> 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 >> > > > _______________________________________________ > 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 >
_______________________________________________ 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