| opendevreview | chandan kumar proposed openstack/watcher-dashboard master: Add Playwright-based E2E testing framework https://review.opendev.org/c/openstack/watcher-dashboard/+/970353 | 04:43 |
|---|---|---|
| opendevreview | chandan kumar proposed openstack/watcher-dashboard master: Add Playwright-based E2E testing framework https://review.opendev.org/c/openstack/watcher-dashboard/+/970353 | 05:14 |
| opendevreview | Alfredo Moralejo proposed openstack/watcher-specs master: Add spec for Action Plan Transformers Framework https://review.opendev.org/c/openstack/watcher-specs/+/994607 | 07:57 |
| opendevreview | chandan kumar proposed openstack/watcher-dashboard master: Add Playwright integration test for skip action workflow https://review.opendev.org/c/openstack/watcher-dashboard/+/976594 | 08:55 |
| chandankumar | amoralej: sean-k-mooney: Hello, On playwright patch, by using tox venv I am able to install playwright and then use venv playwright binary to install deps and broweser binaries, | 09:12 |
| chandankumar | Feel free to take a another look on https://review.opendev.org/c/openstack/watcher-dashboard/+/970353/79 thank you! | 09:12 |
| amoralej | sure, reviewing now | 09:12 |
| chandankumar | ty! | 09:13 |
| amoralej | chandankumar, in the doc doc/source/contributor/playwright-testing.rst , shouldn't be a previous step to install the system deps with sudo before installing the browsers? sudo python -m playwright install-deps | 09:31 |
| opendevreview | David proposed openstack/watcher master: Add three-node setup for non-real-data jobs https://review.opendev.org/c/openstack/watcher/+/995797 | 10:21 |
| opendevreview | David proposed openstack/watcher master: [DNM] Add three-node setup for non-real-data jobs https://review.opendev.org/c/openstack/watcher/+/995797 | 10:22 |
| opendevreview | Alfredo Moralejo proposed openstack/watcher-specs master: Add spec for Action Plan Transformers Framework https://review.opendev.org/c/openstack/watcher-specs/+/994607 | 10:28 |
| opendevreview | David proposed openstack/watcher master: Add three-node setup and enable scope tests https://review.opendev.org/c/openstack/watcher/+/995797 | 10:33 |
| chandankumar | amoralej: yes, correct, I fixed that part | 10:36 |
| opendevreview | chandan kumar proposed openstack/watcher-dashboard master: Add Playwright-based E2E testing framework https://review.opendev.org/c/openstack/watcher-dashboard/+/970353 | 10:36 |
| opendevreview | David proposed openstack/watcher master: [DNM] Add three-node setup and enable scope tests https://review.opendev.org/c/openstack/watcher/+/995797 | 10:37 |
| opendevreview | David proposed openstack/watcher-tempest-plugin master: Add test suite for audit scope https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/995799 | 10:39 |
| opendevreview | David proposed openstack/watcher-tempest-plugin master: [DNM] Add test suite for audit scope https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/995799 | 10:40 |
| opendevreview | Alfredo Moralejo proposed openstack/watcher-specs master: Add spec for Action Plan Transformers Framework https://review.opendev.org/c/openstack/watcher-specs/+/994607 | 11:01 |
| sean-k-mooney | chandankumar: amoralej im inlcidne to say perferct is the enamy of good enough. the playwright patch is "good enough" to merge now and we can likely clean up the other nits seperatly | 11:11 |
| amoralej | wfm | 11:13 |
| chandankumar | sounds good. | 11:17 |
| sean-k-mooney | we have the irc meeting today right? | 11:18 |
| sean-k-mooney | in an hour? or two hours? | 11:18 |
| sean-k-mooney | if doug is happy wiht it we can merg it during the meeting | 11:18 |
| chandankumar | +1 | 11:20 |
| sean-k-mooney | i think the premetion spec is also close | 11:32 |
| sean-k-mooney | there are a few things to refine | 11:32 |
| sean-k-mooney | jgilaber: you raissed some valid point that likely shoudl be adressed | 11:32 |
| sean-k-mooney | i wonder if we realsiticlly can adress all of it by the end of the day | 11:33 |
| sean-k-mooney | today being the spec freeze deadlien we shoudl disscusin the meetingf if we are ok with granting ti an expction till next week or not. | 11:33 |
| sean-k-mooney | i would not want to extended the deadline beyond next weeks irc meeging | 11:34 |
| sean-k-mooney | *meeting | 11:34 |
| dviroel | o/ | 11:57 |
| dviroel | hi all, a reminder that starting today, we have a new meeting time | 11:58 |
| dviroel | https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/message/CLQ7AEJO32LIOQIX5J4YTXVMHQECZOVZ/ | 11:58 |
| dviroel | the meeting will start at 1:00 PM | 11:59 |
| dviroel | feel free to add your topics to the agenda | 11:59 |
| dviroel | https://etherpad.opendev.org/p/openstack-watcher-irc-meeting#L31 | 11:59 |
| sean-k-mooney | dviroel: the only itmes i had was playwright and the permetion spec | 12:00 |
| sean-k-mooney | see the converstaion above | 12:00 |
| dviroel | ack, i was reading back | 12:00 |
| dviroel | giong to revisit both now | 12:01 |
| winiciusallan[m] | o/ | 12:01 |
| winiciusallan[m] | i've just started my day. i'll check the comments in the preemptible spec in a bit | 12:02 |
| sean-k-mooney | no worreies. just a procedural matter today is spec freeze so the topic i have for the meeting is do we enfoce that and derfer your spec to nextg cycle or grant a limited excpetion to say next week to allwo you to fix the remaiing small peices | 12:03 |
| winiciusallan[m] | i'm okay with guys prefer to move this spec for the next cycle as the deadline is today and we have a few things to refine | 12:14 |
| winiciusallan[m] | i had a time requirement on my end and I would need to merge it on this cycle, but I don't have it anymore and I'm okay in following what works better | 12:15 |
| dviroel | winiciusallan[m]: we can discuss in the meeting, no worries | 12:20 |
| opendevreview | Alfredo Moralejo proposed openstack/watcher-specs master: Add spec for Action Plan Transformers Framework https://review.opendev.org/c/openstack/watcher-specs/+/994607 | 12:33 |
| dviroel | sean-k-mooney: amoralej: there is already enough +2 in playwright patch, I think that we can proceed to work in improvements as follow up if needed... | 12:54 |
| dviroel | i also +2 | 12:55 |
| dviroel | w+1 already | 12:55 |
| dviroel | chandankumar: thanks for the hard work there | 12:55 |
| dviroel | folks, watcher meeting will start in 3 minutes - agenda https://etherpad.opendev.org/p/openstack-watcher-irc-meeting#L31 | 12:57 |
| dviroel | #startmeeting watcher | 13:00 |
| opendevmeet | Meeting started Thu Jul 2 13:00:08 2026 UTC and is due to finish in 60 minutes. The chair is dviroel. Information about MeetBot at http://wiki.debian.org/MeetBot. | 13:00 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 13:00 |
| opendevmeet | The meeting name has been set to 'watcher' | 13:00 |
| dviroel | hello everyone o/ - who is around today | 13:00 |
| jgilaber | o/ | 13:01 |
| dviroel | courtesy ping: amoralej sean-k-mooney chandankumar morenod rlandy | 13:02 |
| morenod | o/ | 13:02 |
| amoralej | o/ | 13:02 |
| dviroel | lets strat while folks are joining.. | 13:03 |
| dviroel | let's start with today's meeting agenda | 13:03 |
| rlandy | o/ | 13:03 |
| dviroel | #link https://etherpad.opendev.org/p/openstack-watcher-irc-meeting#L31 (Meeting agenda) | 13:03 |
| dviroel | as usual, feel free to add your own topics to the agenda | 13:03 |
| dviroel | starting with | 13:03 |
| dviroel | #topic Announcements | 13:03 |
| sean-k-mooney | o/ | 13:04 |
| dviroel | so yeah, we reach hibiscus milestone 2 | 13:04 |
| dviroel | #link https://releases.openstack.org/hibiscus/schedule.html | 13:04 |
| dviroel | and with that we reach our deadline for new specs | 13:05 |
| dviroel | we were discussing them past week | 13:05 |
| dviroel | we have a few of them that are open | 13:05 |
| dviroel | #link https://review.opendev.org/q/project:openstack/watcher-specs+is:open | 13:05 |
| dviroel | i think that both Add spec for migration eligibility filters feature and | 13:06 |
| dviroel | Add spec for Action Plan Transformers Framework | 13:06 |
| dviroel | will need to be moved to 2027.1 as we didn't had enough time to discuss and refine | 13:07 |
| dviroel | the first one I'm proposing and i intend to move it to 2027.1 folder | 13:07 |
| winiciusallan[m] | o/ | 13:07 |
| sean-k-mooney | im ok with continuting to review it for 2027.1 but ya with the cyborg spec i did not have time to look at it | 13:07 |
| dviroel | amoralej: any objection? | 13:07 |
| dviroel | we can still discuss the specs yeah, with more time | 13:08 |
| amoralej | would it work if we give one more week to review the transformers? | 13:08 |
| dviroel | do you think that is feasible to implement it in 2026.2? | 13:08 |
| amoralej | anyway, iiuc i can send patches for implementation even if it's not approved yet | 13:09 |
| amoralej | lemme recheck dates | 13:09 |
| sean-k-mooney | i would perfer to instead comple say the funciotnal testing owrk | 13:09 |
| sean-k-mooney | if you have time this cycle then start this work with the other work in flight | 13:10 |
| amoralej | i sent all the patches for functional as i'm aware | 13:10 |
| dviroel | i am reviewing the functional test framework yeah | 13:10 |
| amoralej | more tests may be added but the framework is complete | 13:10 |
| sean-k-mooney | ack i have not been following that sorry | 13:10 |
| dviroel | we can get back to them in the reviews topic, soon | 13:11 |
| jgilaber | I haven't reviewed that either but would the functional framework allow us testing the ironic helper? | 13:11 |
| sean-k-mooney | when you say compelt does that include regression tsts and non gabbit tests | 13:11 |
| amoralej | one thing i can do is to still have patches for the transformers even if it's not approved and get it back on it early after 2026.2 | 13:11 |
| dviroel | i didn't had time to check the Action Plan Transformers framework spec yet, not sure how complex it would be | 13:12 |
| sean-k-mooney | ok looking at i tthe asse seams to be yes | 13:12 |
| amoralej | sean-k-mooney, i'd say so | 13:12 |
| chandankumar | o/ | 13:12 |
| sean-k-mooney | jgilaber: it could | 13:12 |
| amoralej | np dviroel legt's prioritize the other work | 13:12 |
| dviroel | it is not only about imlpementation but also about review bandwidth in the end | 13:13 |
| jgilaber | +1 that is the worry | 13:13 |
| amoralej | jgilaber, i didn't add ironic service fixtures, but may be added following similar approach if that's good | 13:13 |
| sean-k-mooney | my perfernce woudl be to defer ther tramsformes work and destination filterign to next cycle | 13:13 |
| sean-k-mooney | but to start in the implemiton and contionue the review | 13:14 |
| amoralej | ok | 13:14 |
| jgilaber | thanks, I'll try to make some time to review them and might add that later in the cycle | 13:14 |
| sean-k-mooney | baiscly aim to complete them for milestoen 1 | 13:14 |
| dviroel | yeah, I do want to have destination filter proposal merged earlier in the next cycle too... | 13:15 |
| dviroel | we can try that | 13:15 |
| amoralej | also there are some bugs, that i may dedicate some time to | 13:15 |
| dviroel | yeah | 13:15 |
| dviroel | or any other spec less improvement.. | 13:15 |
| dviroel | ack, so the last spec in the list is | 13:15 |
| amoralej | yep, no problem, let's do that | 13:15 |
| dviroel | #link https://review.opendev.org/c/openstack/watcher-specs/+/987171 | 13:15 |
| sean-k-mooney | well blueprint freeze is today as well | 13:15 |
| sean-k-mooney | so bugs or docs changes | 13:16 |
| sean-k-mooney | ci that sort of thing | 13:16 |
| dviroel | there is a lot in LP to be covered yes | 13:16 |
| dviroel | wrt to Add spec for Preemptible Instances feature | 13:17 |
| dviroel | which is feasible to get into 2026.2 | 13:17 |
| dviroel | winiciusallan[m]: is updating the spec periodically | 13:17 |
| winiciusallan[m] | i think we are towards an agreed solution | 13:18 |
| dviroel | but there are still some points that we need to close before merging it | 13:18 |
| winiciusallan[m] | there are some implementation details that we can discuss/refine after the approval | 13:18 |
| winiciusallan[m] | yeah, agree | 13:18 |
| dviroel | we need to at least define the prioritization options and the efficacy indicators | 13:19 |
| dviroel | and it seems that reviewers agree that we don't need to do all in the first implementation | 13:19 |
| dviroel | please correct me if i'm wrong | 13:19 |
| dviroel | winiciusallan[m]: note that, according with the release schedule, code need to be approved and merge at the end of august | 13:20 |
| dviroel | which means that implementation, test and docs need to be proposed eralier | 13:20 |
| winiciusallan[m] | ack dviroel | 13:20 |
| sean-k-mooney | right, to that point i think defering the utilstiaon based aspect woudl limit scope a lot | 13:20 |
| dviroel | so reviewers have time to review them | 13:20 |
| sean-k-mooney | and it somethign we coudl add incrementally next cycle | 13:20 |
| dviroel | ack, same thing for multiple efficacy indicators | 13:21 |
| dviroel | note that this also reduces the amount of code to be reviewed later | 13:21 |
| dviroel | winiciusallan[m]: it would depend if it fits your minimal requirement | 13:21 |
| dviroel | amoralej: jgilaber - any thoughts on the spec? | 13:22 |
| winiciusallan[m] | the spec state in a feature is good for me and i think it is feasible to the end of august | 13:22 |
| winiciusallan[m] | s/in a feature sense | 13:22 |
| dviroel | in order to get it merge, we would likely need to extend the deadline to monday I think | 13:22 |
| amoralej | I think it's close to be merged, I think updates covered most comments. As you said we should agree on the efficacy indicators | 13:23 |
| amoralej | let's extend up to next meeting | 13:24 |
| dviroel | winiciusallan[m]: once we add input_parameters and efficacy indicators, it is harder to deprecated, other than add new ones in the future | 13:24 |
| jgilaber | I think it's reasonable to have an extension for this spec | 13:24 |
| amoralej | (which i will be on pto btw) | 13:24 |
| jgilaber | what is open are not major changes I think | 13:24 |
| sean-k-mooney | dviroel: right once added they are basiclly supproted for ever | 13:24 |
| dviroel | we are already in July, so it is expected that winiciusallan[m] already start implementing it | 13:24 |
| sean-k-mooney | dviroel: removing them would be possibel but we need to think carfuly how to do that | 13:24 |
| dviroel | yeah, that why we are focusing on that part of the spec, winiciusallan[m] | 13:25 |
| sean-k-mooney | winiciusallan[m]: do you think you can reine the spec with the remaining itmes by monday and we will review it and make the final decison next week? | 13:25 |
| winiciusallan[m] | yes! | 13:26 |
| sean-k-mooney | any objections to that plan? | 13:26 |
| sean-k-mooney | if not we can likely move on | 13:26 |
| dviroel | the sooner the better | 13:27 |
| winiciusallan[m] | I will address comments today and you guys can see when have time today/tomorrow | 13:27 |
| winiciusallan[m] | or even next week | 13:27 |
| sean-k-mooney | if its updated today ill review same tomorrow | 13:27 |
| jgilaber | sounds good to me | 13:27 |
| dviroel | ok, no objections | 13:28 |
| dviroel | #agreed extend "Add spec for Preemptible Instances feature" spec deadline to next week | 13:28 |
| dviroel | winiciusallan[m]: thanks, we should agree in the details soon I think | 13:28 |
| dviroel | we can continue discussing them in the spec | 13:28 |
| dviroel | or here in the channel at any time | 13:28 |
| dviroel | next topic then | 13:28 |
| winiciusallan[m] | alright. thanks for extending and keep reviewing | 13:28 |
| dviroel | oh, still in m2 topic | 13:29 |
| dviroel | releases | 13:29 |
| dviroel | jgilaber: we can propose some stable releases if not yet proposed | 13:29 |
| dviroel | we recently merged some bugfixes in 2025.2 right? | 13:29 |
| jgilaber | yes, I meant to look at the git history to check if they are tipically proposed automatically | 13:30 |
| dviroel | we still need to focus on 2025.1 patches | 13:30 |
| jgilaber | if not I'll do it tomorrow | 13:30 |
| jgilaber | we have merged patches in all but 2025.1 I think | 13:30 |
| dviroel | but at least 2025.2 has some important fixes to have | 13:30 |
| dviroel | oh good then | 13:30 |
| dviroel | thanks jgilaber - i can sync with you afterwards then | 13:31 |
| dviroel | next announcement | 13:31 |
| dviroel | quick one | 13:31 |
| dviroel | #topic OpenStack "I" Release Naming Pool | 13:31 |
| dviroel | in case you missed this announce in the openstack ml | 13:31 |
| dviroel | #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/NL3364WME3LMUIJA5NXXJOXOAGP6SDSB/ | 13:31 |
| dviroel | you can vote and help to select next release name | 13:32 |
| dviroel | ok.. | 13:32 |
| dviroel | any other announcement? | 13:32 |
| opendevreview | Merged openstack/watcher-dashboard master: Add Playwright-based E2E testing framework https://review.opendev.org/c/openstack/watcher-dashboard/+/970353 | 13:32 |
| dviroel | ^ yey | 13:33 |
| dviroel | #topic Reviews | 13:33 |
| dviroel | this is something important to discuss today | 13:33 |
| dviroel | we had some patches that require reviewers attention | 13:33 |
| dviroel | and it is expected that more will be proposed in the following weeks | 13:33 |
| dviroel | i would like to see your opinion around how to better track them | 13:34 |
| dviroel | the proposal that I have is to keep the status etherpad updated | 13:34 |
| dviroel | #link https://etherpad.opendev.org/p/watcher-2026.2-status | 13:34 |
| dviroel | there we have our list of active blueprints, features and bugfixes | 13:35 |
| sean-k-mooney | that defintly one option | 13:35 |
| sean-k-mooney | that is useful at the end fo the cycle when creating the release prelude ectra | 13:35 |
| sean-k-mooney | the other is to add a review priroity lable to our projects | 13:36 |
| sean-k-mooney | how that woudl work is when a core does a reivew if they think its close to merge they woudl set review pirioty +2 (we woudl also use that for gate blocking issues of cves) | 13:37 |
| sean-k-mooney | if anyoen thing the patch is now ready for review they can set RP+1 | 13:37 |
| sean-k-mooney | the two are not mutally exclsive | 13:38 |
| dviroel | i think that both can work together yes | 13:38 |
| jgilaber | is that something that we need to configure in gerrit? | 13:38 |
| dviroel | I do like that in the etherpad we can organize per feature | 13:38 |
| sean-k-mooney | yes via the project config repo | 13:38 |
| sean-k-mooney | dviroel: yes i like both for diffent reasons | 13:39 |
| amoralej | yep, i think we can use both | 13:39 |
| sean-k-mooney | having it in gerrirt with a label means i can trivially seach by it | 13:39 |
| dviroel | since not all patches sometimes are organized per topic | 13:39 |
| sean-k-mooney | and have a dashboard for reviews | 13:39 |
| dviroel | i think that we can also try the gerrit label | 13:39 |
| sean-k-mooney | having it itn etherpad lest use organsise it more cleanly | 13:39 |
| sean-k-mooney | ya there is oen other option we coudl consdier | 13:40 |
| amoralej | but not having RP+1 implies is not ready for review? | 13:40 |
| sean-k-mooney | amoralej: no that just nural | 13:40 |
| sean-k-mooney | it might be but the person is not activlly askign for prirotisation | 13:40 |
| sean-k-mooney | the idea with +1 is to allow highigihg patch that have been up for a while with no feedback or where the autor feels like its ready but a core has not weighed in | 13:41 |
| sean-k-mooney | amoralej: so it would be perferct for your fucntional testing patchs | 13:41 |
| sean-k-mooney | once we started lookign at them we might upgrade it to +2 | 13:42 |
| dviroel | ack, I can check how to enable this for watcher projects, if you folks agree | 13:42 |
| amoralej | +1 | 13:43 |
| sean-k-mooney | if we do this can i ask we keep it simpelr then nova and not over promise anything in the lable | 13:43 |
| jgilaber | +1 for me too | 13:43 |
| sean-k-mooney | dviroel: ill find the patches i sued to configure it in nova | 13:43 |
| dviroel | thanks sean-k-mooney | 13:43 |
| dviroel | #agreed enabling RP in watcher project to help on priotize patches that are good to merge and need attention from core reviewers | 13:44 |
| sean-k-mooney | #link https://review.opendev.org/c/openstack/project-config/+/787523 | 13:44 |
| sean-k-mooney | #link https://review.opendev.org/c/openstack/project-config/+/837595 | 13:44 |
| dviroel | nice | 13:45 |
| sean-k-mooney | i dont like the "promise" nameing there | 13:45 |
| sean-k-mooney | but we can figure that out in the review | 13:45 |
| dviroel | ack, thanks, I can check that after the meeting | 13:46 |
| dviroel | anyone want to bring any other review to attention here? | 13:46 |
| dviroel | We have a list of patches in the review etherpad | 13:47 |
| dviroel | we were talking about the functional test infra before | 13:47 |
| dviroel | #link https://review.opendev.org/q/topic:%22blueprint-functional-test-infrastructure%22 | 13:47 |
| dviroel | i am reviewing the emulators part | 13:48 |
| dviroel | and I plan to try them in my env before adding my votes too | 13:48 |
| sean-k-mooney | yep i will continue reviewing that today i have one request in the first patch | 13:48 |
| sean-k-mooney | basiclly fixtures -> local_fixutres or tests/common/fixtures | 13:48 |
| dviroel | ack | 13:49 |
| amoralej | thanks for your reviews, i'll check comments | 13:49 |
| sean-k-mooney | tldr if we call that module fixutre and you configure an ide to find tests under watcher/tests it breaks | 13:49 |
| dviroel | interesting point, i think that amoralej can check that | 13:49 |
| sean-k-mooney | the work around is to confiure the ide to be able to run only the unit test or funcitonal tests at a time | 13:49 |
| dviroel | so, in the end, I think that functional tests may save us lot of additional test that we plan to have in watcher-tempest | 13:50 |
| sean-k-mooney | its easeer ot fix now then later hence the -1 | 13:50 |
| sean-k-mooney | yes | 13:50 |
| dviroel | which are are now taking long to finish | 13:50 |
| amoralej | ok, yep, that seems easy to fix | 13:50 |
| sean-k-mooney | or rather we can still have soem of those tests | 13:50 |
| dviroel | so, as sooner we get the functional infra, the better | 13:51 |
| dviroel | since we should have new patches proposal together with features and bugfixes | 13:51 |
| amoralej | i've been locally using those tests to test other patches, i.e. the openstacksdk ones | 13:51 |
| sean-k-mooney | but the funciotnl test will be a lot cheaper to implemnt edgcases and failure/regressions cases | 13:51 |
| dviroel | which would increase the amount of tests and the time to complete the ci jobs | 13:51 |
| amoralej | and while reviewing the default_parameters one, also may be easily used to validate many cases fast | 13:52 |
| dviroel | we would still need end-to-end tests for features | 13:52 |
| sean-k-mooney | actully so that an interstin gpoint. for nova at least we hae a seperate tox env for the fucntional test to keep the job time shorter | 13:52 |
| dviroel | but maybe not all scenarios need to be tested with tempest | 13:52 |
| amoralej | i agree we can probably clean some from tempest | 13:52 |
| sean-k-mooney | i.e. the unit and fucntional test run in parallel | 13:52 |
| sean-k-mooney | dviroel: as we build our conficndnce in the fucntional tests the time to have tempest vs funcitonal will become more clear | 13:53 |
| dviroel | this is already a huge improvement and something that we can focus in 2026.2 | 13:53 |
| sean-k-mooney | basiclly negitive test i.e. tha tyou cant do somethign are almost never right to od in tempest | 13:53 |
| sean-k-mooney | out side of some very basic api assetions | 13:53 |
| sean-k-mooney | but funcitonal tests are great for that | 13:54 |
| sean-k-mooney | they are also good for anything related to timeing issues or repoducing bugs that have interaction with multipel parts | 13:54 |
| sean-k-mooney | we shoudl almos tnever have tempets test for bugs | 13:54 |
| sean-k-mooney | but once you have funcitoal test lamost every bug shoudl start with a functional reguression test to demonstate teh bug and then prevent a regession | 13:55 |
| sean-k-mooney | ill stop talking now :) amoralej ++ | 13:55 |
| dviroel | since there is no bugs listed in the bugs topic, we can use this last 5 min for this, np | 13:55 |
| dviroel | unless someone want to mention any other review/bug | 13:56 |
| dviroel | amoralej: i guess that part of your your was based on nova's functional test infra? | 13:56 |
| dviroel | the standalone part is something only in watcher? | 13:56 |
| amoralej | yes, nova was the inspiration :) | 13:56 |
| amoralej | yes the standalone is mine | 13:56 |
| dviroel | yeah, i didn't check how is implemented in nova.. | 13:57 |
| amoralej | from my own experience having the option to run in standalone is good | 13:57 |
| dviroel | but I remember that you mentioned | 13:57 |
| dviroel | amoralej: it is, when runnig scaling tests for example, it has been useful for me | 13:57 |
| amoralej | also, i added an option to send logs for functional test to file, which i found useful when executing manually | 13:58 |
| dviroel | ack | 13:58 |
| dviroel | any other comment around that? | 13:59 |
| amoralej | apart from that, the implementation follows the ideas from the spec and nova mostly | 13:59 |
| dviroel | and the reviews topic? | 13:59 |
| amoralej | i have one othe | 13:59 |
| dviroel | sure | 13:59 |
| amoralej | about https://review.opendev.org/c/openstack/watcher/+/994777 , note that that is related to cinder, where jgilaber is also working | 13:59 |
| dviroel | ack | 14:00 |
| amoralej | so I'd propose to merge that ^ so that jgilaber can rebase on top | 14:00 |
| amoralej | unless it requires changes that make us go to the other way around | 14:00 |
| dviroel | ack, maybe we can also highlight in the status etherpad... I will take a look | 14:00 |
| amoralej | it's the equivalent to the change in nova model for zone_migrate | 14:00 |
| amoralej | i added into the bugfixes | 14:00 |
| dviroel | in openstack-sdk topic | 14:01 |
| amoralej | in the etherpad | 14:01 |
| dviroel | thanks amoralej | 14:01 |
| amoralej | good point | 14:01 |
| dviroel | oh, almost forgot: | 14:01 |
| dviroel | #topic Volunteers to chair next week | 14:01 |
| dviroel | i will be offline next week due to a holiday | 14:01 |
| jgilaber | I can do it | 14:01 |
| dviroel | need someone to chair it | 14:01 |
| dviroel | thanks jgilaber | 14:01 |
| dviroel | let's wrap up for today | 14:01 |
| dviroel | we are out of time | 14:01 |
| dviroel | we will meet again next week | 14:02 |
| dviroel | thank you all for participating | 14:02 |
| dviroel | #endmeeting | 14:02 |
| opendevmeet | Meeting ended Thu Jul 2 14:02:15 2026 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 14:02 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/watcher/2026/watcher.2026-07-02-13.00.html | 14:02 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/watcher/2026/watcher.2026-07-02-13.00.txt | 14:02 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/watcher/2026/watcher.2026-07-02-13.00.log.html | 14:02 |
| opendevreview | Joan Gilabert proposed openstack/watcher master: Remove unused CinderHelper methods https://review.opendev.org/c/openstack/watcher/+/994841 | 14:03 |
| opendevreview | Joan Gilabert proposed openstack/watcher master: Add dataclass wrappers for CinderHelper API objects https://review.opendev.org/c/openstack/watcher/+/994842 | 14:03 |
| opendevreview | Joan Gilabert proposed openstack/watcher master: Replace python-cinderclient with openstacksdk https://review.opendev.org/c/openstack/watcher/+/994843 | 14:03 |
| opendevreview | Joan Gilabert proposed openstack/watcher master: Remove python-cinderclient dependency https://review.opendev.org/c/openstack/watcher/+/994844 | 14:03 |
| opendevreview | Joan Gilabert proposed openstack/watcher master: Use real objects in volume migration tests https://review.opendev.org/c/openstack/watcher/+/994862 | 14:03 |
| opendevreview | Alfredo Moralejo proposed openstack/watcher master: Remove Cinder API calls from zone_migration strategy https://review.opendev.org/c/openstack/watcher/+/994778 | 14:36 |
| opendevreview | chandan kumar proposed openstack/watcher-dashboard master: Add Playwright integration test for skip action workflow https://review.opendev.org/c/openstack/watcher-dashboard/+/976594 | 14:47 |
| amoralej | sean-k-mooney, wrt dir names for fixtures, what if i use watcher/tests/local_fixtures for fixtures used both in unit and functional and watcher/tests/functional_fixtures for the ones only for functional? | 15:22 |
| sean-k-mooney | amoralej: that was my orgianl propoasl | 15:28 |
| sean-k-mooney | well not quite | 15:29 |
| sean-k-mooney | i was proposing one for all fixture | 15:29 |
| sean-k-mooney | if you want to split it that is ok too | 15:29 |
| amoralej | it may be simpler to just put everything in one | 15:29 |
| amoralej | to avoid discuss if this is shared or common :) | 15:29 |
| sean-k-mooney | amoralej: basiclly eventlet used to prevent use singel step debuging through functional tests and unit test swithout hack | 15:30 |
| sean-k-mooney | i really like that that now work again with the threaded mode | 15:30 |
| sean-k-mooney | but i dont liek haveign to work aroudn the fixtures directory name issues | 15:30 |
| sean-k-mooney | amoralej: so im fine with whatever name you chose | 15:31 |
| sean-k-mooney | as long as its not "fixtures" | 15:31 |
| amoralej | in https://specs.openstack.org/openstack/watcher-specs/specs/2026.1/approved/functional-testing.html there is local_fixtures and fixtures, that's probably where i took that from, but doing all fixtures i local_fixtures seems the easiest path | 15:31 |
| amoralej | said that, in the spec local_fixtures was the place for the common ones | 15:31 |
| amoralej | and in https://gist.github.com/SeanMooney/43afa55282d2286a312eae7f3c7709e2 only mentions local_fixtures | 15:32 |
| amoralej | so i will do that | 15:32 |
| sean-k-mooney | ya in the gist i had alrady accouneted for this issue | 15:32 |
| amoralej | i'll do that, thanks! | 15:33 |
| sean-k-mooney | cool | 15:33 |
| sean-k-mooney | i ned to quickly write a patch for cyborg for the enfoce socpe thing but ill try to do a fully review of your funtional test work tomowworw | 15:33 |
| sean-k-mooney | dviroel: we shoudl also review watchers policy by the way | 15:33 |
| sean-k-mooney | i wont have a change to do it this week but we should ensure we are usign scopes properly before the end of the release | 15:34 |
| sean-k-mooney | looking very quickly we are not usign them at all | 15:35 |
| sean-k-mooney | so we may break | 15:35 |
| sean-k-mooney | all of our api shoudl be project scope as a baseline | 15:35 |
| sean-k-mooney | we can discuss supprot other scope later but that is the default | 15:36 |
| sean-k-mooney | that means we need to add the scope defition to all our polices | 15:36 |
| sean-k-mooney | gmaan: ^ is there a default scope set in oslo.plocy that woudl be appled here https://github.com/openstack/watcher/blob/master/watcher/common/policies/action.py | 15:37 |
| gmaan | sean-k-mooney: there is no default scope, if no scope_type in policy rule then oslo.policy will skip the scope enforcement | 15:37 |
| sean-k-mooney | ok so we can adress this later in the cycle | 15:38 |
| gmaan | it is mainly driven from policy rule, if scope is expected then oslo.policy check scope of token otherwise not | 15:38 |
| gmaan | sean-k-mooney: yes | 15:38 |
| sean-k-mooney | i have not really tought about if watch shoudl allwo system or doamin scoped tokens to use our api | 15:39 |
| sean-k-mooney | my incliation is it has neve rbeen tested so no by default | 15:39 |
| gmaan | k, yeah you can do it later and enforce_scope removal will not impact the policy who does not have scope yet | 15:39 |
| sean-k-mooney | and default everythign to project and then we can consdier where to go form there | 15:39 |
| gmaan | sounds good | 15:40 |
| sean-k-mooney | dviroel: after i remove the usage in cyrbog i might propose a poc of that for watcher in a week or two when i have time to get back to this | 15:40 |
| sean-k-mooney | we can dicsus how to proceed in one of the irc meetings | 15:41 |
| * dviroel back from lunch, reading back | 16:12 | |
| dviroel | yeah, we can check that in the following weeks | 16:38 |
| dviroel | not sure about system scoped, but project/domain scope I think that would require changes in how watcher manages resources.. | 16:39 |
| winiciusallan[m] | sean-k-mooney: I believe I'll move to change the resource:<name> key to use placement allocation | 16:40 |
| winiciusallan[m] | so we can support passing GPU stuffs or other resources in the future | 16:40 |
| winiciusallan[m] | CUSTOM_NVIDIA_*, ... | 16:41 |
| sean-k-mooney | dviroel: so doamin im not sure about. nova does not supprot domain token for example so we might but we will have to thinking about it | 16:52 |
| sean-k-mooney | winiciusallan[m]: that is ok but the reason i was defering that is that means we need to add the allcoation to the cluster data model for the instnace | 16:53 |
| sean-k-mooney | we can do that, its not hard | 16:53 |
| sean-k-mooney | but you are not allow to make rest api calls to placment durign the execution of the strage during an audit | 16:54 |
| sean-k-mooney | so any data you need about hte intance need to be provie via the cluster data modle | 16:54 |
| sean-k-mooney | it would not be hard to add this but just letting you know the wider context | 16:54 |
| winiciusallan[m] | sean-k-mooney: don't we have the flavor information in the CDM? | 16:55 |
| sean-k-mooney | winiciusallan[m]: we do but that is not the same | 16:55 |
| sean-k-mooney | im not refering to matich agaisnt resouce: flavor extra specs | 16:56 |
| sean-k-mooney | allocation request for resouce come form many diffent locations | 16:56 |
| sean-k-mooney | and direct resouce: usage in the flavor is discuraged | 16:57 |
| sean-k-mooney | the only usage of it today is for the legacy vgpu supprot | 16:57 |
| opendevreview | Winicius Allan Bezerra da Silva proposed openstack/watcher-specs master: Add spec for Preemptible Instances feature https://review.opendev.org/c/openstack/watcher-specs/+/987171 | 17:01 |
| opendevreview | Alfredo Moralejo proposed openstack/watcher master: Move test fixtures to local_fixtures directory https://review.opendev.org/c/openstack/watcher/+/988388 | 18:42 |
| opendevreview | Alfredo Moralejo proposed openstack/watcher master: Fix race condition in DefaultLoader._reload_config() https://review.opendev.org/c/openstack/watcher/+/994845 | 18:42 |
| opendevreview | Alfredo Moralejo proposed openstack/watcher master: Add functional test framework for Watcher https://review.opendev.org/c/openstack/watcher/+/988389 | 18:42 |
| opendevreview | Alfredo Moralejo proposed openstack/watcher master: Add Nova/Placement emulators and host_maintenance functional tests https://review.opendev.org/c/openstack/watcher/+/993352 | 18:42 |
| opendevreview | Alfredo Moralejo proposed openstack/watcher master: Add Prometheus emulator and workload_stabilization functional tests https://review.opendev.org/c/openstack/watcher/+/993543 | 18:42 |
| opendevreview | Alfredo Moralejo proposed openstack/watcher master: Add Cinder emulator and zone_migration functional tests https://review.opendev.org/c/openstack/watcher/+/993634 | 18:42 |
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!