<p>+1</p>
<p>Very excited to see this in and usable</p>
<p>Duncan Thomas</p>
<div class="gmail_quote">On Jan 30, 2015 2:56 AM, "Philipp Marek" <<a href="mailto:philipp.marek@linbit.com">philipp.marek@linbit.com</a>> wrote:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi all,<br>
<br>
> in Paris (and later on, on IRC and the mailing list) I began to ask around<br>
> about providing a DRBD storage driver for Nova.<br>
> This is an alternative to using iSCSI for block storage access, and would<br>
> be especially helpful for backends already using DRBD for replicated<br>
> storage.<br>
any news about this?<br>
<a href="https://review.openstack.org/#/c/134153/" target="_blank">https://review.openstack.org/#/c/134153/</a><br>
<br>
<br>
To reiterate:<br>
* Spec was submitted in time (Nov 13)<br>
* Spec wasn't approved, because the Cinder implementation<br>
(<a href="https://review.openstack.org/#/c/140451/" target="_blank">https://review.openstack.org/#/c/140451/</a>) merge got delayed,<br>
because its prerequisite (<a href="https://review.openstack.org/#/c/135139/" target="_blank">https://review.openstack.org/#/c/135139/</a>,<br>
Transition LVM Driver to use Target Objects) wasn't merged in time,<br>
because on "deadline day" (Dec. 17) Gerrit was so much used that<br>
this devstack run got timeouts against some python site during<br>
setup<br>
* "Spec Freeze Exception" was submitted in time<br>
<a href="http://lists.openstack.org/pipermail/openstack-dev/2015-January/054225.html" target="_blank">http://lists.openstack.org/pipermail/openstack-dev/2015-January/054225.html</a><br>
* Having DRBD for Cinder is good, but using the *same* protocol<br>
to the Nova nodes should really help performance (and reliability);<br>
for example, the transitions<br>
Network -> Kernel -> iSCSI daemon -> Kernel -> Block Device<br>
wouldn't be needed anymore; the Kernel could directly respond to the<br>
queries, and in the near future even using RDMA (where available).<br>
Reliability should be improved as the Nova node can access multiple<br>
storage nodes _at the same time_, so it wouldn't matter if one of them<br>
crashes for whatever reason.<br>
<br>
<br>
Please help us *now* to get the change in. It's only a few lines in<br>
a separate driver, so until it gets configured it won't even be<br>
noticed!<br>
<br>
<br>
And yes, of course we're planning to do CI for that Nova driver, too.<br>
<br>
<br>
Regards,<br>
<br>
Phil<br>
<br>
--<br>
: Ing. Philipp Marek<br>
: LINBIT | Your Way to High Availability<br>
: DRBD/HA support and consulting <a href="http://www.linbit.com" target="_blank">http://www.linbit.com</a> :<br>
<br>
DRBD® and LINBIT® are registered trademarks of LINBIT, Austria.<br>
<br>
__________________________________________________________________________<br>
OpenStack Development Mailing List (not for usage questions)<br>
Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" target="_blank">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
</blockquote></div>