12:00:57 <dviroel> #startmeeting watcher 12:00:57 <opendevmeet> Meeting started Thu Aug 28 12:00:57 2025 UTC and is due to finish in 60 minutes. The chair is dviroel. Information about MeetBot at http://wiki.debian.org/MeetBot. 12:00:57 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 12:00:57 <opendevmeet> The meeting name has been set to 'watcher' 12:01:16 <dviroel> hi all o/ 12:01:17 <chandankumar> \o 12:01:23 <dviroel> who's around today? 12:01:30 <morenod> o/ 12:01:30 <jgilaber> o/ 12:02:07 <dviroel> courtesy ping: amoralej sean-k-mooney 12:02:14 <sean-k-mooney> o/ 12:02:18 <dviroel> amoralej is out today.. 12:02:22 <dviroel> lets start 12:02:32 <dviroel> today's meeting agenda 12:02:48 <dviroel> #link https://etherpad.opendev.org/p/openstack-watcher-irc-meeting#L30 (Meeting agenda) 12:02:53 <dviroel> feel free to add your own topics to the agenda 12:03:03 <dviroel> first one 12:03:27 <dviroel> #topic Flamingo Feature Freeze 12:03:42 <dviroel> we reach the flamingo feature freeze 12:03:53 <dviroel> #link https://releases.openstack.org/flamingo/schedule.html 12:04:11 <dviroel> which means that no feature should be landed after this deadline, unless your explicit require an exception 12:04:33 <dviroel> but I don't think that is the case of watcher this cycle :) 12:04:45 <sean-k-mooney> techinally its end of day 12:04:57 <dviroel> right 12:05:03 <sean-k-mooney> so the only "feautre" i think we may merge woudl be the watcher client update 12:05:09 <sean-k-mooney> for the skip change 12:05:15 <dviroel> if we check our etherpad 12:05:17 <sean-k-mooney> althogh i have not reviewed that properly yet 12:05:19 <dviroel> #link https://etherpad.opendev.org/p/watcher-flamingo-status 12:05:31 <dviroel> Host Maintenance Strategy - Disable Migration [MERGED] 12:05:51 <dviroel> main patch merged, we only have tempest plugin change to merge, which is fine 12:06:07 <dviroel> but requires some attention from the owner 12:06:27 <dviroel> i will send a comment in the patch later today 12:06:33 <sean-k-mooney> yep that can merge at any time once ready 12:06:42 <dviroel> Add Skip Actions [MERGED] 12:07:07 <dviroel> as sean-k-mooney mentioned, we still have the client to review/merge 12:07:14 <dviroel> the main patches landed 12:07:48 <dviroel> I alredy did a review in the client, looks good, just added a comment about default microversion 12:07:54 <sean-k-mooney> i see your +1 on the tempest change and chandankumar is +2 12:08:10 <sean-k-mooney> do we want to proceed with is as is 12:08:12 <dviroel> maybe chandankumar can also take a look and help us with a follow up if needed 12:08:18 <sean-k-mooney> and extend in a folow up? 12:08:31 <dviroel> sean-k-mooney: works for me, not a blocker 12:08:59 <sean-k-mooney> your main concern was adding senario tests in addtion to the api tests right 12:09:07 <dviroel> yes 12:09:16 <dviroel> it is kind of hidden there 12:09:25 <chandankumar> dviroel: I am working on updating python-watcherclient patch 12:09:30 <dviroel> but it is at least validating the implementation 12:09:33 <sean-k-mooney> i woudl like to get some coverage in sooner rather then later so given alfredo will be on pto for a while i think we can proceed with this as a baselien if no one objects 12:09:35 <dviroel> chandankumar: thanks! 12:09:47 <dviroel> sean-k-mooney: agree 12:10:13 <dviroel> we didn't have too much time do discuss that 12:10:27 <dviroel> but it is better to merge as is, and have the tests running 12:10:29 <chandankumar> dviroel: dashboard one is almost ready, I am working on adding status_field to the action info 12:10:51 <dviroel> chandankumar: ok, I haven't look at that yet 12:11:54 <sean-k-mooney> im partly open to the idea of a feature freeze exception if its ready by the end of day tomorrow 12:12:02 <sean-k-mooney> for the dashbaord pices 12:12:14 <sean-k-mooney> but after that i would prefer to adress this next cycle 12:12:23 <sean-k-mooney> that assume we get the clinet change landed today 12:13:00 <dviroel> i am ok with that too 12:13:08 <sean-k-mooney> and can do a relase this week. 12:13:32 <sean-k-mooney> i said this last cycle by the way 12:14:01 <sean-k-mooney> but the general rule of thumb is if its +2w by FF you are allowed to recheck it up until the next monday to get it landed 12:14:23 <sean-k-mooney> after that the grace perod has ended but i dont think we will have that problem 12:14:29 <sean-k-mooney> we dont have a lot of patches pending 12:14:55 <dviroel> chandankumar: ok, so we count on you for updating those, let us know if you need help with something 12:15:28 <opendevreview> chandan kumar proposed openstack/python-watcherclient master: Add action update command to support skipping actions manually https://review.opendev.org/c/openstack/python-watcherclient/+/956911 12:15:41 <dviroel> oh, you are fast :) 12:15:41 <chandankumar> dviroel: ^^ 12:16:31 <dviroel> ok, sean-k-mooney will also need to look at the client and we can see if is really ready 12:16:38 <dviroel> tks chandankumar 12:16:48 <dviroel> I will check the dashboard later 12:17:03 <dviroel> anything else for this feature? 12:17:27 <dviroel> next one is 12:17:29 <dviroel> Extend Compute Model Attributes [MERGED] 12:17:49 <dviroel> we discussed in previous meetings the limitatio of the max api version in the novaclient 12:18:05 <dviroel> which limited us to a smaller set of attributes added in the model 12:18:34 <dviroel> we merged the feature based on that 12:18:50 <dviroel> a spec updated is open for review 12:19:03 <dviroel> #link https://review.opendev.org/c/openstack/watcher-specs/+/955921 12:19:11 <dviroel> which address these concerns 12:19:23 <opendevreview> chandan kumar proposed openstack/watcher-dashboard master: Add option to SKIP Actions https://review.opendev.org/c/openstack/watcher-dashboard/+/958209 12:19:44 <dviroel> with that, the api microversion 1.6 should be the max microversion for this cycle 12:20:14 <dviroel> the changes that we miss to land are the tempest plugin ones too 12:20:36 <dviroel> which cover some tests to data_model list (1.3) 12:20:43 <dviroel> and the extended attributes (1.6) 12:20:56 <dviroel> no changes are expected in the client 12:21:18 <dviroel> the data model list in the client already display a limited list of attributes 12:21:56 <dviroel> only the most important ones like, host and server uuids, states, etc. 12:22:05 <sean-k-mooney> ack, again on the tempest changes we can merge those up to rc1, we done branch the plugin so we can merge after rc 1 as well but we may want to sablise the plugin whiel the final release is happening 12:22:10 <sean-k-mooney> we can dicuss tha as we go 12:22:28 <dviroel> ack 12:22:57 <sean-k-mooney> ya im not entirly sure what to do about this api in general and the client support for it 12:23:11 <sean-k-mooney> this si one of the api that i think was a mistake in the past to expsoe in the rest api 12:23:42 <dviroel> right, and tbh I think that today is only useful for tempest tests 12:23:47 <sean-k-mooney> we coudl continue to supprot it but if we want to ever run more then 1 decsion enghine we need to refactor how we store the datamodel 12:24:22 <sean-k-mooney> so for now im not worreied about the divergace betwen the api respocne and the cli view 12:24:28 <sean-k-mooney> because as you said its already a subset 12:25:03 <dviroel> correct 12:25:21 <sean-k-mooney> but i do think we want to think about this before adding https://review.opendev.org/c/openstack/watcher/+/955365 12:26:05 <jgilaber> yes, as dviroel said, the only use for this is in tempest tests 12:26:09 <sean-k-mooney> basicly do we want to remove it in a new microversoin effectivly freezing the api respocne to what is in 1.6 12:26:12 <dviroel> ack, we need to discuss more about that 12:26:27 <sean-k-mooney> or continue to extned it in which case we need to prioritize making it scaleable 12:26:28 <jgilaber> and notifications solve the problem without it 12:27:44 <dviroel> this may be a good topic for another meeting or ptg 12:27:52 <sean-k-mooney> yep 12:28:05 <sean-k-mooney> ill add it to the ptg list if its not already there 12:28:14 <dviroel> ack, thanks sean-k-mooney 12:28:33 <dviroel> ok, that's it for this feature 12:28:54 <dviroel> the other 2 we already covered in last week's meeting 12:29:11 <dviroel> Eventlet Removal and Aetos datasource are done for this cycle 12:29:22 <dviroel> with CI jobs already in place 12:29:38 <dviroel> anything else that we should discuss about FF? 12:30:27 <sean-k-mooney> on thing 12:30:36 * dviroel waits 12:30:52 <sean-k-mooney> tthe monasca client patch 12:30:57 <dviroel> oh 12:30:59 <sean-k-mooney> https://review.opendev.org/c/openstack/watcher/+/956968 12:31:08 <sean-k-mooney> ill be rebaseing this today 12:31:19 <sean-k-mooney> but i woudl like to treat that as a bugfix not a feature 12:31:42 <sean-k-mooney> which is why i filed abug so assuming there is no objection 12:31:46 <dviroel> ack 12:31:52 <sean-k-mooney> then that can merge up until rc1 12:32:10 <sean-k-mooney> so just callign out that im not expecting this to be blocked by FF unless there i an objection 12:32:25 <sean-k-mooney> i should get it rebased in an hour or so anyway 12:32:30 <sean-k-mooney> but that was all i had 12:32:39 <dviroel> is there any objections on sean-k-mooney proposal? 12:32:45 <jgilaber> +1 from me to consider it a bug, since the project is retired 12:33:20 <sean-k-mooney> i guess there is a followup but ill also need input form the stable team 12:33:39 <sean-k-mooney> the followup being once we are happy with it do we want to backport it or not 12:33:49 <sean-k-mooney> this si not normlaly something we would backport 12:34:05 <sean-k-mooney> but again its now retired and monascal has not had a release since 2023.1 12:34:10 <dviroel> right, it has some upgrade impact 12:34:26 <sean-k-mooney> so i woudl like to get input form the stable team on that 12:34:37 <dviroel> thanks sean-k-mooney - lets us know how it goes 12:34:55 <dviroel> ok, so I will move to the next topic 12:35:13 <dviroel> #topic Elections for PTL/TC are open 12:35:38 <dviroel> a reminder that elections for PTL/TC are open 12:35:56 <dviroel> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/5WLCZVX7VSIVQDWOSH6KTHM62275WIQW/ 12:36:33 <dviroel> if you are eligible to vote, please check your email, you should receive a link to the voting pool 12:37:52 <dviroel> and watcher will go with DPL model once again 12:38:15 <dviroel> the current model proposal is here 12:38:18 <dviroel> #link https://review.opendev.org/c/openstack/governance/+/958085 12:38:49 <dviroel> i think that we are just missing dan approval there 12:38:52 <dviroel> to proceed 12:39:30 <dviroel> I may ping him later today to get his attention to this patch 12:39:45 <dviroel> any question or concerns? 12:40:01 <dviroel> #topic Reviews 12:40:19 <dviroel> in general, i would propose you to add your patches to the review list 12:40:21 <dviroel> https://etherpad.opendev.org/p/watcher-flamingo-status#L41 12:40:37 <dviroel> i already start reviewing some of them from there 12:40:41 <opendevreview> chandan kumar proposed openstack/watcher-dashboard master: Add option to SKIP Actions https://review.opendev.org/c/openstack/watcher-dashboard/+/958209 12:41:02 <dviroel> anyone wants to bring any specific change for review? 12:41:40 <dviroel> ok, as mentioned before, we have until rc1 to merge bugfixes, so the focus next will be these changes 12:41:51 <dviroel> next topic 12:42:05 <dviroel> #topic Bug Triage 12:42:11 <dviroel> #link https://bugs.launchpad.net/watcher/+bug/2121147 (Storage model assumes 'total_volumes' is reported by all drivers) 12:42:18 <dviroel> already covered last week 12:42:28 <dviroel> but we are missing some triage there 12:43:12 <dviroel> jgilaber: it doesn't affect any strategy for now right? 12:43:27 <jgilaber> I think that can be considered low/medium, it does not cause the audits to fail 12:43:35 <jgilaber> just pollute the logs 12:43:38 <dviroel> ack, i would vote for low then 12:44:25 <dviroel> jgilaber: feel free to upgrade for medium if you find that it affect more than initially expected 12:44:35 <dviroel> next is 12:44:35 <sean-k-mooney> ya if it does not fail i woudl not set it above medium 12:44:36 <jgilaber> ack 12:44:42 <sean-k-mooney> low sound ok 12:44:43 <dviroel> #link https://bugs.launchpad.net/watcher/+bug/2120666 (Errors appear in decision-engine logs when cinder, notifications and storage collector are enabled) 12:44:52 <dviroel> also covered last week 12:45:29 <jgilaber> this is similar imo 12:45:55 <jgilaber> the model builds find, just logs an exception without enough context 12:45:58 <dviroel> not sure if it affects how the model is built 12:46:12 <jgilaber> I meant to say the model builds fine 12:46:20 <dviroel> or if just ignore the error and continues 12:46:22 <dviroel> ack 12:46:30 <dviroel> so we set to Low as the previous 12:46:44 <jgilaber> +1 12:46:44 <dviroel> if we identify that it affects more, we increase the importance 12:47:01 <dviroel> nice 12:47:04 <dviroel> next 12:47:10 <dviroel> #link https://bugs.launchpad.net/watcher/+bug/2121384 (Watcher docs have an incorrect way of configuring cinder notifications) 12:47:23 <dviroel> this is a new one 12:47:38 <dviroel> thanks jgilaber for reporting the issue 12:47:39 <sean-k-mooney> yep this came out of a conversation i had with jgilaber 12:47:47 <jgilaber> I opened this one when I tried to follow to docs to enable cinder notifications 12:48:01 <dviroel> so this is already in progress, and it is a doc update right 12:48:07 <jgilaber> I improved the patch with suggestions from sean-k-mooney 12:48:25 <sean-k-mooney> the patch is not linked to the bug by hte way 12:48:28 <jgilaber> now it also fixes the default values for the notification topics that watcher listens to 12:48:50 <jgilaber> #link https://review.opendev.org/c/openstack/watcher/+/957668 12:49:03 <jgilaber> ^this is the patch, not sure why is not being linked to the bug 12:49:29 <dviroel> oh, so it is not doc only 12:49:50 <jgilaber> initially I thought it would be doc only, but it grew slightly in scope 12:50:02 <jgilaber> the patch itself is really small though 12:50:19 <sean-k-mooney> so it should have a release note 12:50:28 <sean-k-mooney> but im debating 12:50:39 <sean-k-mooney> if we want to keep listing to the watcher topic as well 12:50:42 <sean-k-mooney> or replace it as you have 12:50:54 <sean-k-mooney> long term the replacement is correct 12:51:05 <jgilaber> ack, I'll add the release note 12:51:12 <sean-k-mooney> short term its a trivial but still precense upgrade impact 12:51:25 <opendevreview> chandan kumar proposed openstack/python-watcherclient master: Add action update command to support skipping actions manually https://review.opendev.org/c/openstack/python-watcherclient/+/956911 12:51:41 <sean-k-mooney> what i would personally do is have 2 patches 12:51:45 <dviroel> jgilaber: i think that because needs to be "Closes-Bug: #1234567" 12:51:48 <sean-k-mooney> the fisrt woudl jsut add the new topic 12:51:50 <jgilaber> I'm fine adding the openstack.notification and removing the wtacher one in next cycle 12:51:57 <sean-k-mooney> the second would remove the old one 12:52:08 <sean-k-mooney> we would backprot the first patch but not the second 12:52:28 <dviroel> +1 12:52:37 <jgilaber> ok, sounds good, I can do that 12:52:48 <sean-k-mooney> cool 12:53:00 <dviroel> should we stay with that bug as Low? 12:53:02 <sean-k-mooney> so this is a docs/defaulitng but 12:53:15 <sean-k-mooney> *bug but you can workaround it manually so ya low 12:53:17 <jgilaber> dviroel, I can try that but other patches like https://review.opendev.org/c/openstack/watcher/+/951323/15//COMMIT_MSG worked without the '#' 12:53:32 <dviroel> jgilaber: yeah right, it was a guess 12:53:52 <sean-k-mooney> the bot might be picker 12:53:57 <dviroel> yep 12:54:07 <sean-k-mooney> the link rednerign in gerrit is sperate form the bot that adds the links ot launchpad 12:54:32 <jgilaber> maybe it's because of the docs tag? 12:54:43 <sean-k-mooney> no that wont affect anything 12:55:05 <sean-k-mooney> the other posiblity is its becasue you didnt add it in v1 12:55:20 <sean-k-mooney> in any case it tivial to add a liknk as a comment in the bug 12:55:22 <sean-k-mooney> but try the # 12:55:26 <dviroel> yeah 12:55:39 <sean-k-mooney> if that does not work can you just add the link in a comment on lauhcpad 12:55:49 <jgilaber> yes, I think that's right, I forgot it in the first PS 12:55:50 <dviroel> ok, anything else on this bug? otherwise I will move to the next topic 12:56:07 <jgilaber> all good for me 12:56:08 <dviroel> #topic release stuff 12:56:21 <dviroel> sean-k-mooney: want to highlight that one? 12:56:26 <sean-k-mooney> ya this will be short 12:56:31 <dviroel> #link https://review.opendev.org/c/openstack/releases/+/958468 12:56:39 <sean-k-mooney> so i just summerised what need to get done in the next 3 weeks 12:56:52 <sean-k-mooney> so today we releaed all the stable branches with the security fix 12:57:12 <sean-k-mooney> next week we need to start workign on the cycle highlight and release note prelude 12:57:28 <sean-k-mooney> the week after we prepare RC1 12:57:35 <opendevreview> Merged openstack/watcher-tempest-plugin master: Add api test for skip action https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/955775 12:57:51 <sean-k-mooney> and once that is relased master will reopen for 2026.1 12:58:05 <sean-k-mooney> at rc1 stable 2025.2 will be created 12:58:15 <sean-k-mooney> and there are a few bot propospoed patches we need to merge 12:58:42 <sean-k-mooney> so if folks want to knwo more i put more links in the ether pad 12:58:52 <dviroel> nice, thanks sean-k-mooney 12:59:33 <dviroel> thanks for the release liason work during these last cycles 12:59:33 <chandankumar> sean-k-mooney: can you summarize realse tasks for next release in a etherpad so that Joan and I can pick up. 12:59:43 <jgilaber> thanks for the info sean-k-mooney! 12:59:48 <dviroel> chandankumar: https://docs.openstack.org/watcher/latest/contributor/release-guide.html 13:00:00 <dviroel> there is a lot of info here too ^ 13:00:12 <sean-k-mooney> chandankumar: ya so we are at https://docs.openstack.org/watcher/latest/contributor/release-guide.html#week-following-milestone-3 13:00:17 <chandankumar> dviroel: thanks! 13:00:28 <sean-k-mooney> and we need to complete the next few sections in order 13:00:46 <sean-k-mooney> after RC1 ships we start the hand of to the new release liasons 13:01:03 <dviroel> chandankumar: jgilaber: if you find that some info is missing, we can update this official doc then 13:01:17 <jgilaber> ack, thanks 13:01:30 <chandankumar> thanks! 13:02:00 <dviroel> jgilaber: chandankumar tks for taking this role 13:02:02 <sean-k-mooney> by the way we have to do this for all our delibvier not just watcher core 13:02:18 <sean-k-mooney> i.e. the release note prelue ectra although we only really need 1 cycle highlishts 13:02:39 <jgilaber> that's the client and dashboard but not the tempest plugin, correct? 13:02:53 <sean-k-mooney> mostly correct 13:02:55 <sean-k-mooney> https://docs.openstack.org/watcher/latest/contributor/release-guide.html#project-context 13:03:20 <sean-k-mooney> we do not branch the tempest plugin and its release with the other tempest plugins following there timeline 13:03:30 * dviroel time check 13:03:43 <sean-k-mooney> and watcher-specs is not released at all 13:03:45 <jgilaber> ah, right 13:03:49 <sean-k-mooney> but ya read over that doc it has the details 13:04:13 <dviroel> thanks morenod for volunteering for next meeting chair 13:04:20 <dviroel> let's wrap up for today 13:04:27 <dviroel> thank you all for participating 13:04:34 <dviroel> #endmeeting