09:00:14 <priteau> #startmeeting blazar
09:00:15 <openstack> Meeting started Tue Feb  5 09:00:14 2019 UTC and is due to finish in 60 minutes.  The chair is priteau. Information about MeetBot at http://wiki.debian.org/MeetBot.
09:00:16 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
09:00:18 <openstack> The meeting name has been set to 'blazar'
09:00:48 <priteau> #topic Roll call
09:01:11 <masahito> o/
09:01:19 <priteau> Hi masahito
09:02:38 <priteau> Is tetsuro_ around as well?
09:02:45 <tetsuro_> 0/
09:02:50 <priteau> Hi tetsuro_
09:02:55 <tetsuro_> Hi
09:03:42 <priteau> #topic stein-3 milestone
09:04:05 <priteau> We're getting closer to the stein-3 milestone which is at the end of February
09:04:38 <priteau> Actually, it moved, it is now March 7
09:05:20 <priteau> We should try and merge as much as possible before then
09:05:49 <priteau> I've reviewed the floating IP patch set and left various comments
09:06:28 <masahito> Thanks your review. I checked your comments.
09:06:38 <masahito> I'll soon update it.
09:06:42 <priteau> Thanks masahito
09:06:54 <priteau> I've also started to revisit bug fixes from Chameleon, but I ran into a placement-related issue testing them
09:07:17 <priteau> #link https://bugs.launchpad.net/blazar/+bug/1814594
09:07:18 <openstack> Launchpad bug 1814594 in Blazar "Failure to create a host" [Undecided,New] - Assigned to Tetsuro Nakamura (tetsuro0907)
09:09:49 <tetsuro_> I'll look into that. Could you add some more explanation to reproduce the bug, not only traceback?
09:09:58 <priteau> I am not sure why we haven't seen that in our gate jobs. This is on a DevStack deployed yesterday.
09:11:01 <priteau> I just added steps to reproduce
09:11:09 <priteau> 1) deploy DevStack with blazar plugin
09:11:09 <priteau> 2) `source openrc admin admin`
09:11:09 <priteau> 3) blazar host-create <hostname>
09:11:55 <tetsuro_> Okay, I will see that asap.
09:11:58 <priteau> I ran the same thing last week and it worked fine at the time, so it could be an upstream issue with nova or placement
09:13:21 <priteau> I will try submitting a basic patch to see if gate has the same problems
09:14:44 <tetsuro_> That's a good idea. Anyway I'll check the IRC openstack-blazar tomorrow morning.
09:14:49 <priteau> Check https://review.openstack.org/#/c/634890/ in an hour or so. If Zuul fails, it's probably an issue with an upstream change
09:15:17 <priteau> Or tomorrow. I forgot it was already late for you :-)
09:15:42 <tetsuro_> That's okay. I've put myself in as a reviewer of that patch.
09:16:57 <priteau> Next on my list is to review tetsuro_'s affinity patches.
09:17:33 <tetsuro_> Regarding to placement-related bugs, we also need more eye on https://review.openstack.org/#/q/topic:bug/1813252.
09:18:45 <priteau> OK, will check
09:19:38 <tetsuro_> thanks
09:22:10 <priteau> masahito: Did you have a chance to look at tests for allocation API? https://review.openstack.org/#/c/586859/
09:23:01 <masahito> I'm working on the patch.
09:23:28 <masahito> Looks I didn't figure out gabbits itself.
09:24:02 <masahito> I try to understand gabbits well again, then I'll update it.
09:25:25 <priteau> OK. Let me know if you need some help.
09:26:03 <masahito> thanks
09:27:23 <priteau> Anything else to discuss for stein-3?
09:29:49 <priteau> If not, let's move to AOB
09:29:51 <priteau> #topic AOB
09:30:51 <priteau> If you've missed it, there was an interesting discussion on the mailing list about reservation of bare-metal nodes
09:30:53 <priteau> #link http://lists.openstack.org/pipermail/openstack-discuss/2019-January/002223.html
09:31:15 <priteau> Among the features they wanted, it included:
09:31:22 <priteau> - leasing: a mechanism for marking nodes as available, requesting nodes for a specific length of time, and returning those nodes to the available pool when a lease has expired.
09:31:42 <priteau> Of course I advised them to use Blazar for this purpose :-)
09:32:20 <priteau> So it's possible that in the future we will have a baremetal plugin
09:32:55 <masahito> I saw the interesting discussion, too :-)
09:33:45 <masahito> Is the baremetal support Chameleon the different approach?
09:34:13 <priteau> They
09:34:22 <priteau> They are referring to different approaches
09:35:03 <priteau> In Chameleon's case, Nova is used as the interface to provision Ironic nodes. There is some adaptation needed only because host aggregates don't work for Ironic.
09:36:09 <priteau> In the scenario discussed on the ML, Nova is not involved, and even Ironic is much less involved than normal. They want to reserve nodes for some time, and be able to deploy them with their own baremetal provisioner, which could be something else than Ironic (e.g. Foreman)
09:37:01 <priteau> It will require some changes to Ironic, or an additional proxy service above it, to be implemented first
09:38:23 <masahito> I see.
09:39:17 <priteau> It would be amazing if it brings new contributors to Blazar.
09:39:44 <masahito> right.
09:40:01 <priteau> We need to make sure Blazar doesn't have bad bugs when they try it!
09:40:08 <masahito> And he/she has real usecases. It's nice.
09:40:15 <priteau> Anything else to discuss for AOB?
09:40:46 <priteau> If not I would suggest an easy code review: https://review.openstack.org/#/c/628469/
09:42:30 <priteau> It's just little syntax changes to make the new hacking happy
09:47:11 <priteau> I forgot another thing to mention
09:47:20 <priteau> tetsuro_: I added you to blazar-core on Launchpad
09:47:52 <priteau> I also gave admin privileges to all currently active cores
09:47:56 <tetsuro_> Thank you!
09:48:28 <priteau> I think that's all for today then. Have a good rest of the week!
09:48:46 <priteau> #endmeeting