15:00:46 <mattcrees[m]> #startmeeting blazar 15:00:46 <opendevmeet> Meeting started Thu Jun 12 15:00:46 2025 UTC and is due to finish in 60 minutes. The chair is mattcrees[m]. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:46 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:46 <opendevmeet> The meeting name has been set to 'blazar' 15:00:56 <mattcrees[m]> Hello o/ 15:01:11 <markpowers> o/ 15:01:49 <mattcrees[m]> #topic Flamingo cycle 15:02:42 <priteau> o/ 15:03:06 <mattcrees[m]> How are things looking on this cycle for you two, any patches to draw attention to? 15:05:19 <priteau> We need to review the various patches submitted by StackHPC colleagues. I haven't been able to get much time on Blazar recently. 15:06:22 <mattcrees[m]> I haven't had much time either unfortunately. I've seen this patch looks good to merge imo https://review.opendev.org/c/openstack/blazar/+/937239, but the others need a closer look 15:06:27 <markpowers> I don't have patches prepared yet, but we did deploy the flavor plugin. 15:07:57 <mattcrees[m]> That's good to hear, has it been working ok for you? 15:10:02 <markpowers> Yes, there was a bug where instances from flavor reservations were detected as preemptible, and deleted once another lease on the same host starts. I didn't have a chance to confirm this happens with the upstream code, but it seemed to be the case at first glance 15:12:23 <mattcrees[m]> Ouch, sounds pretty serious. 15:13:31 <priteau> Preemptibles is a preview feature really 15:15:11 <markpowers> this add_computehost call 15:15:29 <markpowers> the add_comphost call will remove all running instances on a host https://opendev.org/openstack/blazar/src/branch/master/blazar/plugins/instances/instance_plugin.py#L616-L617 15:16:38 <markpowers> fortunately we caught it before any users were impacted, but as a workaround I reverted the preemptible commit 15:17:10 <opendevreview> Pierre Riteau proposed openstack/blazar master: Remove support for Python 3.8 and 3.9 https://review.opendev.org/c/openstack/blazar/+/952471 15:17:59 <priteau> I documented it here: https://docs.openstack.org/blazar/latest/user/architecture.html 15:18:03 <priteau> "Only the compute host reservation is compatible with preemptible instances at the moment." 15:18:23 <priteau> But we should fix it ;) 15:19:39 <mattcrees[m]> Perhaps all the while it's not compatible, we shouldn't allow use of the instance reservations if preemptibles are enabled? 15:19:59 <markpowers> I'll check our configuration, I didn't think we enabled it 15:20:36 <opendevreview> Pierre Riteau proposed openstack/blazar-dashboard master: Remove support for Python 3.9 https://review.opendev.org/c/openstack/blazar-dashboard/+/952472 15:23:18 <opendevreview> Pierre Riteau proposed openstack/python-blazarclient master: Remove support for Python 3.8 and 3.9 https://review.opendev.org/c/openstack/python-blazarclient/+/952473 15:24:08 <opendevreview> Pierre Riteau proposed openstack/blazar-nova master: Remove support for Python 3.8 and 3.9 https://review.opendev.org/c/openstack/blazar-nova/+/952474 15:26:33 <priteau> Just submitted a few spring cleaning patches for the Flamingo cycle 15:27:36 <mattcrees[m]> Thanks priteau, I'll take a look at these. 15:30:29 <mattcrees[m]> One other thing I had to raise: I've seen the October 2025 PTG team signup is now open. I'll get us signed up again 15:30:37 <priteau> I wanted to raise this topic 15:31:01 <priteau> PTG is October 27-31, 2025 15:32:36 <mattcrees[m]> I'm only working the 27th and 30th that week, but we usually do Thursday so hopefully that won't be an issue for scheduling. 15:33:14 <priteau> October 30 could work 15:35:07 <markpowers> that works for me 15:36:05 <mattcrees[m]> Cool I'll sign us up then :) 15:36:37 <mattcrees[m]> We can confirm on the exact time slot once booking opens closer to the time 15:37:20 <mattcrees[m]> That's it for me this week, was there anything else either of you would like to discuss? 15:39:19 <markpowers> nothing from me 15:40:09 <priteau> Nothing from me 15:40:29 <mattcrees[m]> Ok, thanks for joining both :) 15:40:39 <mattcrees[m]> #endmeeting