| opendevreview | Ivan Anfimov proposed openstack/openstack-manuals master: install-guide: Add 2026.1 Gazpacho release https://review.opendev.org/c/openstack/openstack-manuals/+/983534 | 11:09 |
|---|---|---|
| opendevreview | Ivan Anfimov proposed openstack/openstack-manuals master: install-guide: Add 2026.1 Gazpacho release https://review.opendev.org/c/openstack/openstack-manuals/+/983534 | 11:11 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-manuals master: install-guide: Add 2026.1 (Gazpacho) release https://review.opendev.org/c/openstack/openstack-manuals/+/983534 | 11:12 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-manuals master: Community support - update until 2026.1 (Gazpacho) https://review.opendev.org/c/openstack/openstack-manuals/+/983536 | 11:14 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-manuals master: Community support - update until 2026.1 (Gazpacho) https://review.opendev.org/c/openstack/openstack-manuals/+/983536 | 11:16 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-manuals master: Installation Guide - Overview and Environment - add 2026.1 (Gazpacho) release https://review.opendev.org/c/openstack/openstack-manuals/+/983537 | 11:17 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-manuals master: Installation Guide - Environment - add 2026.1 (Gazpacho) release https://review.opendev.org/c/openstack/openstack-manuals/+/983537 | 11:20 |
| gouthamr | tc-members: gentle reminder that the weekly IRC meeting will be hosted here in ~55 minutes | 16:02 |
| gouthamr | Agenda is here: https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee#Agenda | 16:02 |
| gouthamr | #startmeeting tc | 17:00 |
| opendevmeet | Meeting started Tue Apr 7 17:00:37 2026 UTC and is due to finish in 60 minutes. The chair is gouthamr. Information about MeetBot at http://wiki.debian.org/MeetBot. | 17:00 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 17:00 |
| opendevmeet | The meeting name has been set to 'tc' | 17:00 |
| gouthamr | Welcome to the weekly meeting of the OpenStack Technical Committee. A reminder that this meeting is held under the OpenInfra Code of Conduct available at https://openinfra.dev/legal/code-of-conduct. | 17:01 |
| gouthamr | Today's meeting agenda can be found at https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee | 17:01 |
| dansmith | o/ | 17:01 |
| gouthamr | #topic Roll Call | 17:01 |
| noonedeadpunk | o/ | 17:01 |
| frickler | \o | 17:01 |
| gouthamr | courtesy-ping: cardoe, mnasiadka, bauzas | 17:02 |
| gouthamr | noted absence: s p o t z | 17:02 |
| dansmith | I _think_ bauzas is out today | 17:02 |
| gouthamr | oh, ack dansmith | 17:03 |
| mnasiadka | O/ | 17:03 |
| -opendevstatus- NOTICE: Load on the opendev.org Gitea backends is under control again for now, if any Zuul jobs failed with SSL errors or disconnects reaching the service prior to 16:15 UTC they can be safely rechecked | 17:04 | |
| gouthamr | fungi: i don't know if this is from the earlier status attempt? ^ | 17:05 |
| cardoe | I'm here. Just once again squirreled in other channels. | 17:05 |
| gouthamr | let's get started | 17:06 |
| gouthamr | #topic Last week's AIs | 17:06 |
| gouthamr | noonedeadpunk had one on venus/vitrage retirement/deprecation. The patches are here: | 17:07 |
| gouthamr | #link https://review.opendev.org/c/openstack/governance/+/981959 (Venus) | 17:07 |
| gouthamr | #link https://review.opendev.org/c/openstack/governance/+/982869 (Vitrage) | 17:07 |
| gouthamr | noonedeadpunk: missing the ML notification for this though | 17:07 |
| gouthamr | os-monasca repo retirement is blocked by a project-config change: | 17:09 |
| gouthamr | #link https://review.opendev.org/c/openstack/project-config/+/974773 | 17:09 |
| gouthamr | maybe frickler and fungi can take a look again when they're available | 17:09 |
| gouthamr | think requestsexceptions is in the same boat: | 17:10 |
| gouthamr | #link https://review.opendev.org/c/openstack/project-config/+/979808 (Retire requestsexceptions) | 17:10 |
| frickler | yes, I have that on my list | 17:10 |
| noonedeadpunk | Damn,yes | 17:10 |
| noonedeadpunk | I wanted to push, but then easter | 17:10 |
| noonedeadpunk | Will do this today for sure | 17:11 |
| gouthamr | no problem, we'll likely want to give folks a couple of weeks to respond after you've done that and proceed with the patches | 17:11 |
| gouthamr | venus' retirement will need the repos to be emptied out as well.. | 17:11 |
| gouthamr | i actually don't know how the deprecation would work, but i guess similar? i.e., vitrage repos' master branch should have all of it's content removed | 17:12 |
| fungi | #link https://docs.openstack.org/project-team-guide/repository.html#deprecating-a-repository Deprecating a Repository | 17:13 |
| fungi | (and if that's wrong/broken somehow, please help update the instructions there) | 17:13 |
| gouthamr | ty for sharing that fungi | 17:14 |
| gouthamr | on to the next one: | 17:15 |
| gouthamr | #link https://review.opendev.org/c/openstack/governance/+/981832 (Gerrit hashtags proposal) - we discussed this a bit more last week, this one needs some votes | 17:15 |
| gouthamr | finally, a couple of AIs that'll be mixed up with announcements/news: | 17:16 |
| gouthamr | the coordinated release of 2026.1 is past us, cycle-trailing projects will have some time to work with this release now: | 17:16 |
| gouthamr | #link https://www.youtube.com/watch?v=YSrqj-pKBPo (Introducing OpenStack 2026.1 'Gazpacho') | 17:16 |
| cardoe | Ironic is doing one this Thursday. | 17:17 |
| gouthamr | we had some lowlights: AI crawler activity overwhelmed opendev.org gitea servers, causing lots of job failures. opendev folks force-approved the "Mark 2026.1 as released" patch to bypass failing tests | 17:17 |
| gouthamr | this issue continues to plague us | 17:18 |
| dansmith | s/us/everyone/ | 17:18 |
| gouthamr | cardoe: ack | 17:19 |
| gouthamr | #link https://openinfra.org/live/ (OpenInfra Live website, next eposode is April 09 @ 14:00 UTC) | 17:19 |
| fungi | yeah, all the communities i'm involved in (not just openinfra-related) are overwhelmed trying to keep up with the llm-training crawler scourge | 17:19 |
| clarkb | one thing to note is that on the CI side you can avoid this entirely by using the zuul git cache | 17:19 |
| clarkb | there is no actual need to use gitea in the Ci system. Its just an easy shortcut and when these problems occur we remember why the intent isfor Zuul to supply all the git info | 17:20 |
| clarkb | (that doesn't fix gitea for everything else though) | 17:20 |
| cardoe | Any instructions on how I can check if stuff is doing that? | 17:20 |
| fungi | if you have example failed builds we can help you track down where it's connecting to opendev.org git instead of using the local cache | 17:20 |
| clarkb | cardoe: I think the primary issue is that global constraints is loaded from the https://opendev.org url rather than the git repo supplied by zuul. its possible many/most/all of those jobs don't have teh requirements repo configured in them as well | 17:21 |
| fungi | feel free to bring up examples in the #openstack-infra channel | 17:21 |
| mnasiadka | fungi: I’ve seen launchpad was also semi-unresponsive - well, still is | 17:21 |
| frickler | yes, they seem to have similar struggles | 17:22 |
| fungi | right, if you hang out in their matrix room, it's been coming up frequently | 17:22 |
| gouthamr | maybe zuul can update the constraints url at run time? | 17:22 |
| fungi | gouthamr: most jobs get the constraints file from the local cache of the requirements repo on job nodes. it's just jobs that are missing it as a required-projects addition that don't, i think | 17:23 |
| fungi | (the standard jobs all inherit from a job that includes it already) | 17:24 |
| gouthamr | ack, these are jobs that aren't based off of the standard ones.. | 17:24 |
| fungi | usually ad hoc tox jobs where the tox.ini specifies to pull from releases.openstack.org when there's no provided local copy | 17:25 |
| fungi | (and that then redirects to gitea) | 17:25 |
| gouthamr | ack, if you know of one to use as an example, we can let project maintainers find more and fix these up | 17:27 |
| fungi | yep, that's why i asked people bring us example failures in #openstack-infra | 17:28 |
| gouthamr | ++ ty | 17:28 |
| gouthamr | alright, the final hanging item on our action items was to find a new meeting time | 17:29 |
| gouthamr | #link https://app.rallly.co/poll/yycMVUsXThVC | 17:30 |
| gouthamr | most of you filled this one, ty for doing it.. | 17:30 |
| gouthamr | tonyb and bauzas didn't | 17:30 |
| cardoe | I did wanna toss something on the tail end in open time. | 17:30 |
| gouthamr | ack | 17:30 |
| gouthamr | i think this is the only time that currently works for folks | 17:31 |
| gouthamr | so we'll keep it | 17:31 |
| gouthamr | i am thankful that you have this on your calendar and you make the time to attend it.. so i can understand the gravity this slot will have in comparison to others | 17:32 |
| gouthamr | that's all the AIs and the side bar on CI etc | 17:33 |
| gouthamr | was anyone else working on something to mention here? | 17:33 |
| cardoe | Well I wanted to mention the MariaDB collation item. | 17:33 |
| gouthamr | cardoe: sure.. | 17:34 |
| cardoe | They stuck with utf8mb3 but they switched to utf8mb3_uca1400_ai_ci being the default. Not sure how that'll impact us or if we want to update any docs. | 17:34 |
| cardoe | noonedeadpunk: would be happy to work with you on a cross-project effort if we need to. | 17:34 |
| noonedeadpunk | This kinda impacts upgrades yes | 17:35 |
| noonedeadpunk | And I actually have not yet decided if we want or not mandate charset in migrations | 17:35 |
| clarkb | who is they in this context? But also utf8mb3 is slated for release at some point in the future (just a reminder I tried to use this as motivation to switch to utf8mb4 rather than do two big db collation migrations) | 17:35 |
| noonedeadpunk | Mariadb | 17:36 |
| clarkb | oh got it this is just a reference to the upstream default | 17:36 |
| fungi | clarkb: s/release/removal/? | 17:36 |
| noonedeadpunk | Yup | 17:36 |
| clarkb | fungi: yes sorry. My typing is terrible today. utf8mb3 will be removed from mysql and mariadb at some unknown point in the future | 17:36 |
| noonedeadpunk | s/mandate charset/mandate collation/ | 17:36 |
| clarkb | so that only utf8mb4 is an option | 17:36 |
| fungi | cool, just making sure i understood what you were trying to say | 17:37 |
| cardoe | clarkb: https://docs.openstack.org/install-guide/environment-sql-database-ubuntu.html so those instructions are what I'm referring to. | 17:37 |
| noonedeadpunk | Yes, right, and I think that having charset should be stated in migrations in a way | 17:37 |
| cardoe | MariaDB 11.4 was the previous LTS (shipped with Ubuntu 24.04). While MariaDB 11.8 is the current LTS (shipping with Ubuntu 26.04) | 17:37 |
| noonedeadpunk | And upgrade of services can make your db very weird | 17:38 |
| cardoe | With the same config file under the two OSes in the former you'll have utf8_general_ci as your collation and in the latter you'll have utf8mb3_uca1400_ai_ci as your collation. | 17:38 |
| noonedeadpunk | ++ | 17:38 |
| cardoe | They previously were going to change utf8 to alias from utf8mb3 -> utf8mb4. But they've pushed that off for now. | 17:38 |
| cardoe | So it seems like we should update our docs to not use an aliased name that's going to change. | 17:39 |
| noonedeadpunk | And we totally need to sort out at least the alias thing and make it consistent throughout services | 17:39 |
| cardoe | Yep | 17:39 |
| noonedeadpunk | The problem is it's not only docs | 17:39 |
| clarkb | yup I guess the primary concern with this transition is that you'll end up with changing behavior afte an upgrade or even inconsistent behavior between services during an upgrade. Making it consistent makes sense to me | 17:39 |
| noonedeadpunk | Half of projects have charset also defined as an alias in their schema | 17:40 |
| noonedeadpunk | And other half don't have it at all | 17:40 |
| cardoe | Right cause you can have DB wide, table wide, and column collation. | 17:40 |
| cardoe | And a bunch of projects use a hodge podge of that in their schema. | 17:40 |
| gouthamr | i like the idea of calling out the problems, drafting something about the migrations before the PTG | 17:41 |
| dansmith | are you arguing that things will break if the collation switches? | 17:41 |
| dansmith | and also.. does the collation change require actually rewriting tables and things? I guess I thought that was more of a behavior-of-queries thing and not a where/how-data-is-stored thing | 17:42 |
| cardoe | So our rally runs on 24.04 are 100%. Our rally runs on 26.04 are like 99%. And it's DB errors when it goes sideways. | 17:42 |
| cardoe | I haven't had anyone look into it but I just remembered the collation change so I'm just bringing it up. | 17:43 |
| noonedeadpunk | Yes, if there's a db upgrade , new tables may be created with a different collation | 17:43 |
| clarkb | dansmith: I think you can change collation without migrating the database beacuse as you say its a query order behavior. It might be possible that some indexes/queries are negatively affected performance wise | 17:43 |
| noonedeadpunk | And if they contain foreign keys, such migrations will just fail | 17:44 |
| dansmith | clarkb: sure, but I'd expect that any (correctness) problems that stem from the collation change were latent bugs anyway | 17:44 |
| noonedeadpunk | We're running 11.8 for 2 cycles now, and we had to revert the change of defaults for mariadb setup to make things work | 17:44 |
| dansmith | noonedeadpunk: can you qualify "make things work" | 17:45 |
| noonedeadpunk | So when table A and table B have different collations, but they depend on each other, engine really can't compare data | 17:45 |
| cardoe | noonedeadpunk: I'd be curious what ya changed. We're just using rally to generate load. It's running against a devstack. So that's why we've not looked into the issue. I wonder if I threw whatever change you did at the Ubuntu 26.04 image if the success rate would go back up. | 17:46 |
| clarkb | dansmith: not necessarily. If things are expected to be case sensitive vs not that is something collation will affect but there is usually one correct answer for a set of data (I don't think case sensitivity is the issue ehre though) | 17:46 |
| noonedeadpunk | So failing alembic migrations is best outcome | 17:46 |
| noonedeadpunk | As they're possible hidden issues as well, on queries | 17:46 |
| dansmith | clarkb: I guess I expect case-sensitive issues to be mostly "clients were assuming the old behavior" problems | 17:47 |
| dansmith | noonedeadpunk: you're talking about if we have two tables keyed to each other using, say, UUID strings the comparisons will fail? | 17:47 |
| cardoe | Anyway, just something I wanted to point out. If anyone's interested in helping come up with how we should clean up and ensure we are consistent. I'm volunteering noonedeadpunk to help. And anyone else, we'll just follow up separately and report back if there's any project wide goals that need to happen. Make sense gouthamr? | 17:48 |
| noonedeadpunk | uuids will fail on alembic attempt to upgrade schema, as uuids are mostly defined as foreign keys | 17:48 |
| dansmith | either way, I guess I'm not sure what the problem is if an operator just switches the collation of all the tables/databases | 17:48 |
| noonedeadpunk | oh, of operator switches, there's no problem | 17:48 |
| dansmith | either way, I guess I'm not sure what the problem is if an operator just switches the collation of all the tables/databases | 17:49 |
| noonedeadpunk | it just should be concious decision and they need to be aware they have to change existing schema on OS upgrader | 17:49 |
| dansmith | oops | 17:49 |
| dansmith | isn't the collation something they chose (or defaulted) at setup time independent of our migrations? | 17:49 |
| gouthamr | yeah, time check on this topic.. i feel like we need a structured discussion for this.. maybe continue this discussion after the meeting and draft something up on what should the cross project goal (if any) should be? | 17:49 |
| dansmith | yeah, I guess my point is, this seems like something the operator would handle when upgrading mariadb and reading _their_ release notes, not ours | 17:50 |
| cardoe | gouthamr: yes exactly. | 17:50 |
| noonedeadpunk | right, but if they setup on 24.04, and then upgrade 26.04, and upgrade openstack - then there will be a problem | 17:50 |
| gouthamr | and do add a topic to the next week's agenda if you'd like to bring something up here | 17:50 |
| gouthamr | #topic 2026.2/Hibiscus PTG | 17:50 |
| gouthamr | #link https://etherpad.opendev.org/p/apr2026-ptg-os-tc (TC PTG Etherpad) | 17:50 |
| noonedeadpunk | unless they go and either revert defaults for mariadb, or change colelction to new defaults | 17:50 |
| cardoe | The other item I wanted to mention is previously I think mnasiadka mentioned adding a WSGI server into our PTI. | 17:50 |
| gouthamr | this pad's been filling up slowly | 17:51 |
| gouthamr | i wanted to call out a couple of things, that we've pursued in the past PTGs to some degree of success | 17:51 |
| cardoe | I'm not sure if we want to do that or not. We (OpenStack) are very centered around uWSGI which is in minimal maintenance mode. Many services can switch over to gunicorn (or others) just fine. But some services cannot. | 17:51 |
| gouthamr | cardoe: maybe hold that thought for Open Discussion? | 17:51 |
| cardoe | Yes sorry. | 17:52 |
| gouthamr | back on what the TC's done in the past: | 17:52 |
| gouthamr | 1) Encouraging cross project session scheduling earlier in the week | 17:52 |
| gouthamr | we've seen this was helpful.. any objections doing this again? | 17:52 |
| gouthamr | i wanted PTLs/moderators to talk now and schedule cross project topics in the first couple of days | 17:53 |
| gouthamr | best effort ofcourse | 17:53 |
| gouthamr | the problem we had in the past couple of PTGs is that we didn't send out this memo or people didn't realize this pattern | 17:53 |
| gouthamr | okay, sounds like no objections.. | 17:55 |
| * gouthamr or more like, why are we "discussing" this | 17:55 | |
| gouthamr | lol | 17:56 |
| gouthamr | 2) Reducing length of Monday's TC sessions | 17:56 |
| gouthamr | in the spirit of above talks, I'm hoping to reduce the TC's time to 2 hours on 20th April: 1600-1800 UTC.. | 17:56 |
| gouthamr | This should give project teams more time to do cross project discussions.. i'm hoping that "TC goal" discussions can take up that hour that we're giving up | 17:56 |
| gouthamr | in separate cross project sessions | 17:56 |
| gouthamr | any objections here? | 17:56 |
| gouthamr | so for Apr 20th, i'm hoping to: | 17:57 |
| gouthamr | 3) Focus on two big ticket things on Monday | 17:57 |
| gouthamr | - Community Leaders' forum (a catch all for all the things we need to discuss cross-project, and hear PTL/liaisons' concerns) | 17:57 |
| gouthamr | - A forum with Operators - this stuff is new | 17:57 |
| gouthamr | a while ago, we had a big push for operators' presence in the PTG.. some things worked at the time, so, i was hoping we can try this again.. we'll give operators 40-60 mins to come brainstorm with the TC (and the community) on their high level concerns.. | 17:59 |
| gouthamr | the idea isn't novel, but, the biggest problem i see is moderating this sorta thing | 17:59 |
| gouthamr | ... and we're at the hour | 18:00 |
| gouthamr | i'll pause here, but continue adding stuff after we wrap up the meeting | 18:01 |
| gouthamr | do we want to note anything else for the meeting today? | 18:01 |
| gouthamr | cardoe: ack on the uWSGI concern, this came up a few times.. i think it's a good thing to have a proposal (how's the gunicorn adoption going with project teams? what are the gotchas? do we know if anyone's looking to make things easier for testing with it?) | 18:03 |
| mharley[m] | Hello, yes. 😊 | 18:03 |
| gouthamr | cardoe: maybe add it to the topics next week? | 18:03 |
| cardoe | Will do. | 18:03 |
| gouthamr | mharley[m]: we're in garbage time, but do share if you have anything | 18:04 |
| mharley[m] | I’d like to propose post-quantum cryptography as a cross-project topic that should be discussed during the next PTG. | 18:04 |
| mnasiadka | I don’t know if we need to discuss it a lot, maybe just reflect that devstack uses uWSGI for testing for now (in PTI), and discuss what needs to be done next to test gunicorn, granian or whatever else | 18:04 |
| sean-k-mooney | uwsgi does not really need to be in the pti | 18:05 |
| dansmith | I have pretty low motivation to legislate a WSGI container in the PTI | 18:05 |
| sean-k-mooney | but perhaps a refence that project shoudl not assume a specific wsgi server is better | 18:06 |
| gouthamr | ^ agree | 18:06 |
| mharley[m] | I’m not asking the TC to approve implementations or algorithm choices today. I’m asking for support to make sure this topic is actually addressed at the PTG, because that is the right forum for early cross-project discussion, alignment, and planning. | 18:06 |
| sean-k-mooney | mharley[m]: you can freely propsoe topic for each team or if a popup team si created have a single comunity event | 18:06 |
| mharley[m] | I’ve already submitted a popup team proposal and raised the topic on the mailing list. My request today is that the TC help frame PQC as a valid PTG discussion topic and encourage the relevant teams to participate. | 18:07 |
| fungi | mharley[m]: i would recommend organizing that as part of the already proposed pop-up team for the pqc effort | 18:07 |
| fungi | if you need a separate track for it, let me know and i can add it to the track list for you | 18:07 |
| mharley[m] | @fungi, yes, that would be great. | 18:07 |
| mharley[m] | A useful PTG outcome would be to identify impacted areas, map dependencies, and agree on possible next steps. | 18:07 |
| gouthamr | mharley[m] maybe also volunteer as the TC liaison? i think we were waiting on that one to vote on the pop up team's proposal | 18:08 |
| mharley[m] | You mean, myself? | 18:09 |
| gouthamr | mharley[m]: yes, or anyone else.. we did ask here, and we didn't have any TC members that wanted to volunteer | 18:09 |
| gouthamr | a TC liaison doesn't have to be in the TC, just accessible to it | 18:10 |
| mharley[m] | You have one now. 😊 | 18:10 |
| gouthamr | alright, 10 mins past, sorry to keep this running.. I'll follow up on the PTG plans over the ML today.. if you have suggestions on how the operator forum should go, please do let me know | 18:10 |
| gouthamr | #endmeeting | 18:10 |
| opendevmeet | Meeting ended Tue Apr 7 18:10:25 2026 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 18:10 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/tc/2026/tc.2026-04-07-17.00.html | 18:10 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/tc/2026/tc.2026-04-07-17.00.txt | 18:10 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/tc/2026/tc.2026-04-07-17.00.log.html | 18:10 |
| mharley[m] | I’m crafting a presentation to run during the PTG to disclose more details, including what we’ve done at Red Hat. | 18:11 |
| sean-k-mooney | the ptg is not really a presentation type event although having some content to talk over is perhaps useful | 18:11 |
| fungi | mharley[m]: it sounds like you want a conference type talk venue? the ptg is more for working sessions | 18:12 |
| sean-k-mooney | on the topic of the db charterset and coalation type that need discoaion on a per project basis by the way | 18:12 |
| fungi | mharley[m]: also hopefully you're already coordinating with your colleagues at rh who were bringing up the need for a pop-up team and possible project-wide goal | 18:13 |
| sean-k-mooney | most if not all project i workon expect our api to be case sensitive and utf8mb4 woudl be more approate then utf8mb3 for them too | 18:13 |
| mharley[m] | No, folks. The presentation is more to help going throughout the script. A few slides. | 18:13 |
| gouthamr | you can try, but ideally, use etherpad to introduce the problem and take us through your plan.. so folks can comment inline and you can brainstorm | 18:14 |
| sean-k-mooney | mharley[m]: any issues identifed from those scripts shoudl likely be filed as bugs or feature requests in the relevent tackers but that somethign the popup team member likely shoudl be doing in concert with teh service teams | 18:15 |
| cardoe | sean-k-mooney: unfortunately a few projects have a hard dependency on uWSGI behaviors. | 18:16 |
| sean-k-mooney | cardoe: most are tested with uwsgi and apache mod_wsgi | 18:17 |
| sean-k-mooney | at least they were up until reltivly recently | 18:17 |
| dansmith | cardoe: be sure to distinguish between being able to utilize specific WSGI container features and an actual _dependency_ there | 18:17 |
| sean-k-mooney | but ya i woudl gernally confire that a bug if its not protable to be fixed | 18:17 |
| dansmith | for example, glance can use uwsgi stuff, but works fine without it | 18:17 |
| mharley[m] | By script I don’t mean code. I’m talking about the event script. 😊 | 18:17 |
| cardoe | dansmith: I am. Will not work at all without uWSGI. | 18:17 |
| cardoe | despite what the docs say | 18:18 |
| gouthamr | cardoe: that sounds like a bug, and not the goal of the WSGI standardization effort here wanted: https://governance.openstack.org/tc/goals/completed/pike/deploy-api-in-wsgi.html | 18:18 |
| mharley[m] | fungi: Will you be able to create the track for PQC and let me know? | 18:20 |
| fungi | mharley[m]: sure, i'm just wrapping up some vmt stuff now that we've got the gitea servers under control, will get to it shortly | 18:20 |
| * fungi is wearing all the hats today | 18:21 | |
| cardoe | My point was we've effectively made a defacto standard in devstack on uWSGI when projects exclude apache's mod_uwsgi since they test against one implementation. | 18:21 |
| dansmith | AFAIK, we (redhat) run all our WSGI-using services in mod_wsgi (I assume that's what you meant).. can you name and shame a project you're referring to just for an example to chase? | 18:23 |
| sean-k-mooney | devstack used ot be configurable betwwen aparch and uwsgi. kolla still is although the change there default form apach to uwsgi in the last 2 ish year but yes we use apache in most of oru downstream operators | 18:30 |
| sean-k-mooney | i belive a few are usign uwsgi but nova is still runign under apache for sure | 18:30 |
| dansmith | sean-k-mooney: I didn't think we even had uwsgi available to us there | 18:30 |
| sean-k-mooney | i think that a reltively recent thing unless im miss rememebring | 18:31 |
| dansmith | okay I wasn't aware of any such change, but surely possible | 18:31 |
| sean-k-mooney | i dont think we have it in our nova contianer but i belvie one of the other serives moved to it let me check | 18:31 |
| sean-k-mooney | ah maybe that never merged https://github.com/openstack-k8s-operators/tcib/pull/282 it was for designate | 18:32 |
| dansmith | yeah | 18:33 |
| mharley[m] | Thank you, fungi! | 18:33 |
| * gouthamr has updated the agenda for next week, please feel free to add topics: | 20:39 | |
| * gouthamr https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee | 20:39 | |
| opendevreview | Goutham Pacha Ravi proposed openstack/openstack-manuals master: Documentation main page - add link to Glossary https://review.opendev.org/c/openstack/openstack-manuals/+/947058 | 21:23 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-manuals master: Documentation main page - add link to Glossary https://review.opendev.org/c/openstack/openstack-manuals/+/947058 | 21:25 |
| opendevreview | Goutham Pacha Ravi proposed openstack/governance master: Classify gov changes by hashtag, not topic https://review.opendev.org/c/openstack/governance/+/981832 | 21:37 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-manuals master: Drop retired project (Monasca) https://review.opendev.org/c/openstack/openstack-manuals/+/983649 | 21:39 |
| gouthamr | fungi: you may have received a notification regarding the transfer of https://launchpad.net/~mistral-drivers under https://launchpad.net/~openstack-admins | 21:40 |
| gouthamr | may need to add amorin to it so he can further clean it up, and subscribe more people | 21:41 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-manuals master: Drop retired project (Monasca) https://review.opendev.org/c/openstack/openstack-manuals/+/983649 | 21:41 |
| gouthamr | (cc: JayF who noticed this issue: https://answers.launchpad.net/launchpad/+question/824018) | 21:41 |
| clarkb | I don't see email to our shred email address so maybe it went to fungi directly | 21:42 |
| clarkb | gouthamr: I do note that puppet forge is warning the puppet senlin module will go into "SEEKING CONTRIBUTIONS" state on their platform in october if there aren't any updates made to it before then | 21:42 |
| clarkb | sorry the module name is openstack-senlin on puppet forge | 21:43 |
| gouthamr | oh, we retired that project: https://review.opendev.org/q/topic:%22retire-senlin%22 | 21:45 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-manuals master: Drop retired project (solum) https://review.opendev.org/c/openstack/openstack-manuals/+/983651 | 21:45 |
| gouthamr | clarkb: do we manage anything on forge.puppet? maybe tkajinam knows if we needed any cleanup there in the past | 21:45 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-manuals master: Drop retired project (solum) https://review.opendev.org/c/openstack/openstack-manuals/+/983651 | 21:47 |
| clarkb | gouthamr: https://forge.puppet.com/modules/openstack/senlin/readme I think this is the module which says it is owned by https://forge.puppet.com/modules/openstack which has published things as recently as last october | 21:48 |
| clarkb | so I think that is a maybe yes. But it isn't clear to me if we're still doing so etc | 21:48 |
| gouthamr | october 2021? | 21:48 |
| clarkb | october 2025 per my second link | 21:49 |
| clarkb | october 27, 2025 to be more specific according to the listing of modules there | 21:49 |
| gouthamr | ah, ack.. | 21:49 |
| gouthamr | https://opendev.org/openstack/project-config/src/branch/master/playbooks/publish/puppetforge.yaml#L12 | 21:50 |
| gouthamr | this is doing it, and by design for all puppet modules.. but, i think retired modules should be cleaned up with a manual process: https://help.puppet.com/core/current/Content/PuppetCore/modules_deprecate.htm | 21:51 |
| gouthamr | looks like we are missing "steps" in our retirement for this one too.. it's an external system, like PyPi is | 21:52 |
| gouthamr | maybe tkajinam will tell us if he's explored this in the past.. | 21:52 |
| fungi | gouthamr: i don't see any notifications, but the group does indeed seem to be owned by openstack-admins now, so i can subscribe and set as admin anyone you want me to | 21:54 |
| gouthamr | fungi: amorin :) | 21:54 |
| gouthamr | fungi: likely the notification got eaten by whatever is killing launchpad today | 21:55 |
| fungi | oh, quite likely | 21:55 |
| fungi | amorin is now a member and administrator in https://launchpad.net/~mistral-drivers/+members | 21:56 |
| gouthamr | great ty fungi | 22:33 |
| fungi | of course | 22:41 |
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!