09:00:00 <priteau> #startmeeting blazar
09:00:01 <openstack> Meeting started Tue Sep 10 09:00:00 2019 UTC and is due to finish in 60 minutes.  The chair is priteau. Information about MeetBot at http://wiki.debian.org/MeetBot.
09:00:02 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
09:00:04 <openstack> The meeting name has been set to 'blazar'
09:00:09 <priteau> #topic Roll call
09:00:28 <masahito_> o/
09:00:32 <priteau> Hi masahito_
09:03:19 <priteau> Neither Tetsuro nor Bertrand are online
09:04:12 <priteau> I don't see Asmita either
09:04:17 <priteau> Agenda for today:
09:04:20 <priteau> 1. Shanghai Summit: Forum Submissions
09:04:24 <priteau> 2. Removal of API v2
09:04:26 <priteau> 3. Train release schedule
09:04:29 <priteau> 4. Code review priorities
09:04:31 <priteau> 5. AOB
09:04:53 <priteau> #topic Shanghai Summit: Forum Submissions
09:05:18 <priteau> The call for Forum topics is open, until Sep 20.
09:05:32 <priteau> Would you like to submit a topic?
09:06:41 <masahito_> I don't have any idea for Forum right now.
09:07:42 <priteau> We already have plenty of things on the backlog that we can prototype / implement (preemptibles, quota integration, etc.)
09:09:28 <priteau> I'll ask again next week, in case Tetsuro has ideas
09:10:32 <priteau> #topic Removal of API v2
09:11:22 <priteau> We've deprecated the v2 API several releases ago. One of the patches from Asmita reminded me that the code is still there and I was thinking of deleting it
09:11:33 <priteau> I would probably do it earlier in the U cycle
09:11:36 <priteau> Any thoughts?
09:12:02 <masahito_> +1.
09:13:22 <priteau> I'll propose a patch, but let's focus on Train first
09:14:30 <priteau> #topic Train release schedule
09:14:39 <priteau> #link https://releases.openstack.org/train/schedule.html
09:15:11 <priteau> This week is Train-3 milestone, Feature freeze, Final release for client libraries
09:16:39 <priteau> I'd like to merge all the patches from Asmita
09:17:05 <priteau> I've also been working on a new patch for floating IP reservation update, based on Jacob's
09:17:47 <priteau> There are updates that are tricky to handle with the current code, for example when `required_floatingips` are added
09:18:38 <priteau> Maybe we'll just need to deny updates of some fields
09:18:52 <priteau> Hi tetsuro
09:20:20 <priteau> masahito_: tetsuro: I would appreciate a review of this client patch: https://review.opendev.org/#/c/680989/
09:20:36 <tetsuro> Sorry for being late.
09:22:25 <priteau> tetsuro: We've mentioned Forum topics but didn't have any idea. Is there something you want to propose to discuss in Shanghai?
09:22:42 <tetsuro> Not really.
09:23:45 <priteau> OK, we won't submit any topics.
09:24:42 <priteau> For the Train release, I was just saying that I've been working on a new patch for floating IP reservation update, based on Jacob's
09:25:02 <priteau> It still has some limitations but fixes several bugs
09:26:00 <priteau> In the process I discovered issues with the client for supporting required_floatingips
09:26:06 <priteau> Patch: https://review.opendev.org/#/c/680989/
09:27:10 <priteau> I have also another pending patch to handle lease-update in the same way
09:27:55 <priteau> This week is our last chance for a blazarclient release
09:28:06 <tetsuro> Ack, will review it shortly
09:29:22 <priteau> Thanks
09:29:49 <priteau> I am afraid we won't have time to review network reservation fully as the spec hasn't been approved yet.
09:30:24 <priteau> I don't really want to include this feature without significant testing
09:31:36 <priteau> Better to make sure floating IP reservation is well supported
09:34:35 <priteau> tetsuro: I discussed with Jacob your new patches for optimizing DB queries. He wasn't sure they would help much in their environment because apparently it was fetching allocations which took time.
09:35:41 <priteau> Maybe we could write a script to populate a DevStack DB with lots of allocations, to make this problem reproducible.
09:36:05 <tetsuro> I submitted two patch for that optimization, and was none of them helpful?
09:36:27 <priteau> He's just back from holiday, I asked if he could test them
09:37:26 <tetsuro> okay.
09:37:51 <tetsuro> But yes, having a lot of allocations in devstack sounds a good idea.
09:39:02 <priteau> I'll continue to do more reviews this week
09:39:40 <tetsuro> What I wonder is even with https://review.opendev.org/#/c/679935/ it doesn't help, or he just tried https://review.opendev.org/#/c/678755/ and found it did not help
09:40:51 <priteau> Let's see if he can test both this week
09:41:58 <tetsuro> I'll have a raw SQL scripts for this that can help you debug
09:42:47 <priteau> Did you extract the queries generated by SQLAlchemy?
09:43:27 <tetsuro> I will. It's easy.
09:44:37 <priteau> Thanks, that would be really useful. It's purely read-only queries, so we can run them safely at the MySQL prompt.
09:44:56 <tetsuro> right
09:47:00 <priteau> Anything else to discuss about Train?
09:48:38 <priteau> #topic Code review priorities
09:48:50 <priteau> We've covered this a lot already. Priorities are:
09:49:00 <priteau> 1) client patches since this week is the last chance for the release
09:49:33 <priteau> 2) blazar service patches: mostly Asmita's patches and DB optimization patches
09:50:29 <priteau> Anything else?
09:52:30 <priteau> #topic AOB
09:52:38 <priteau> Anything else to share?
09:53:14 <masahito_> Is the room for Blazar PTG fixed?
09:53:57 <priteau> We haven't heard back about rooms at the PTG. I believe most groups will be in a shared space anyway.
09:54:09 <masahito_> I'm trying to book a hotel and flight for the PTG. So if possible, I want to know the day and time.
09:54:20 <masahito_> okay. Thanks.
09:55:21 <priteau> There will be project update during the summit and project onboarding during the PTG
09:56:15 <priteau> Project update slots are not scheduled yet
09:56:52 <priteau> I can send email to ask when they will be communicated?
09:59:37 <priteau> masahito_: ^
10:00:15 <masahito_> sorry. If possible, please send me once they announce.
10:00:55 <priteau> OK
10:01:03 <priteau> We're out of time, thanks for joining!
10:01:05 <priteau> #endmeeting