Thursday, 2025-08-21

opendevreviewchandan kumar proposed openstack/watcher-dashboard master: Add support for passing parameters while creating audit  https://review.opendev.org/c/openstack/watcher-dashboard/+/95753504:59
opendevreviewchandan kumar proposed openstack/watcher-dashboard master: Add support for passing parameters while creating audit  https://review.opendev.org/c/openstack/watcher-dashboard/+/95753505:02
opendevreviewchandan kumar proposed openstack/watcher-tempest-plugin master: Check for single instance migration to backup node  https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/95817410:45
opendevreviewchandan kumar proposed openstack/watcher-tempest-plugin master: Check for single instance migration to backup node  https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/95817411:42
opendevreviewMerged openstack/watcher-tempest-plugin master: Add custom flavor and dynamic threshold to workload_balance tests  https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/95385311:52
amoralejey12:03
amoralejit's meeting time!12:04
jgilabero/12:04
amoralejfeel free to add your topics to https://etherpad.opendev.org/p/openstack-watcher-irc-meeting12:04
morenodo/12:05
amoralej#startmeeting Watcher IRC meeting, August 21st, 202512:05
opendevmeetMeeting started Thu Aug 21 12:05:22 2025 UTC and is due to finish in 60 minutes.  The chair is amoralej. Information about MeetBot at http://wiki.debian.org/MeetBot.12:05
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.12:05
opendevmeetThe meeting name has been set to 'watcher_irc_meeting__august_21st__2025'12:05
rlandyo/12:05
dviroelhi o/12:06
amoralejcourtesy ping: sean-k-mooney chandankumar 12:06
chandankumaro/12:06
amoralejlet's start with the first topic12:07
amoralejI will go throug all the features work in https://etherpad.opendev.org/p/watcher-flamingo-status#L1312:07
amoralej#topic (Flamingo features) Host Maintenance Strategy - Disable Migration12:08
sean-k-mooneyo/12:08
amoralej#link https://review.opendev.org/q/topic:%22bp//host-maintenance-strategy-disable-migration%2212:08
dviroeljust a reminder that, feature freeze is next week, so these feature need to merge to get into 2025.212:08
amoralejso for that implementation has already a +2 https://review.opendev.org/c/openstack/watcher/+/95253812:08
amoralejso, I assume pending on core review12:09
dviroelyeah, we discussed a lot in the change12:09
amoralejyep12:09
dviroellooks good to merge I think12:09
jgilaber+112:09
dviroelsean-k-mooney also +1 in past reviews12:10
amoralejfor the tempest tests, https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/954214 some changes have been requested although it's in good shape12:10
sean-k-mooneyyes i was waitign for dviroel  to re review12:10
sean-k-mooneyi wanted to finish lookign at the tempest test but i will likely approve it later today since dviroel is now happy with it too12:10
dviroelwe may need some improvements/fixes in the tempest patch, but that will not block the main patch12:11
sean-k-mooneythere are a number of follow up dicssion we shoudl have at teh ptg but i tihnk its in a good postion to move forward12:11
amoralejyes, a lot of good questions appeared on that12:11
sean-k-mooneydviroel: ya i saw your comments about the assertions12:11
amoralejstatus is properly tracked in the reviews, so let's move to next one if that's fine for you12:11
sean-k-mooneyand i agree with them in genral but also that it shoudl nto block the feature form landing12:11
dviroelsean-k-mooney: also, I think that we need to block the test to run on stable branches12:11
dviroelotherwise it will break them12:12
amoralejright ^ 12:12
dviroelbut I need to check that12:12
sean-k-mooneydviroel: yes the test will ned to be skiped on stable we shoudl do this via a tempet plugin config option12:12
sean-k-mooneydefault it to false but set it to true on master12:12
amoralejupstream we only run api tests in stable releases, but if running scenarios ones, it would fail12:12
dviroelack, since there is no new api microversion for them12:12
sean-k-mooneyamoralej: that something we also need to fix12:13
dviroelamoralej: yeah, right!12:13
sean-k-mooneywe shoudl be runnign the senario tests on stable too12:13
amoralejyeah, we discussed already i +1 to that12:13
sean-k-mooneybut that seperate 12:13
amoralejyep12:13
amoralejnext feature?12:13
sean-k-mooneyso ill keep the patch open in a table and loop back to it after the meeting but yes we can move on12:14
dviroel+1 to move12:14
amoralej#topic (Flamingo features) Add Skip Actions12:14
amoralej#link https://review.opendev.org/q/topic:%22blueprint-add-skip-actions%2212:14
amoralejthanks for all the reviews there, jgilaber and dviroel 12:14
amoraleji got reviews for most of the implementation ones and did some fixes12:15
dviroeli am missing the review in the client and finishing the API one, but looks good so far12:15
amoralejwrt question in https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/95577512:15
amoralejtbh, I'm not sure if the test i added should be in the scenario instead of api12:16
dviroelah yeah, I have that one open to reply12:16
dviroelcorrect, the test is testing the functionallity, I missed that12:16
amoraleji execute an entire workflow using dummy12:16
dviroelmainly because we don't usually test the funcionality in api tests I think12:16
amoralejso, maybe it should be an scenario? i saw similar ones in api, that's why i added there, but i was unsure12:16
amoralejI'll wait for your feedback12:17
dviroelbut I was able to check in the logs that your feature is working12:17
amoralejwrt the watcher-dashboard change, it relies on watcherclient patch https://review.opendev.org/c/openstack/python-watcherclient/+/95691112:17
sean-k-mooneyya so the merge order will be watcher then client then dashbaord12:18
amoralejso, i was thinking in waiting for it to be merged and released so that i can update the minimal version of the client12:18
dviroeland client release is next week already12:19
amoralejbut I'm afraid that will be after feature freezee12:19
amoralejwdyt? would that be backportable if we miss this release?12:19
sean-k-mooneywe can do an early release of the client before FF12:19
sean-k-mooneyamoralej: no12:19
sean-k-mooneybut we may allow it to merge in the RC period12:19
amoralejthat'd be good12:20
sean-k-mooneyif we get the feature landed today/tomrrow we can aim for a client release monday and proceed with the dashbaord before FF12:20
amoralejunfortunatelly, I'm going to be out next weeks, will be hard to make, unless someone is willing to take care12:21
amoralejI have a wip patch locally, but still needs some work12:21
sean-k-mooneyack we can see how it plays out12:22
amoralejack, let's see if i can send something today12:22
sean-k-mooneyits not the end fo the world it this land early next cycle in the horizon plugin12:22
sean-k-mooneythe only thing that we need in the plugin is the ablity to mark someting as skiped premetivly right12:23
amoralejand show status_message12:23
chandankumaramoralej: If you get wip of watcher-dashboard, I will iterate over that since currently I am already working in that area12:23
sean-k-mooneyah yes but they coudl see the skipped state when its auto skipped12:23
amoralejyes12:23
amoralejactually, when skipping manually, i'm presenting a form to ask for an optional reason that will be added to the status_message12:24
amoralejthanks chandankumar, let's sync later12:24
sean-k-mooneywe use the release with intermediary release cycle for watcher12:24
sean-k-mooneyso we can also do an early release next cycle if we chosoe too12:25
amoralejok, sounds good12:25
amoralejanything else wrt this feature?12:25
amoralej#topic (Flamingo features) Extend Compute Model Attributes12:26
amoralej#link https://review.opendev.org/q/topic:%22bp/extend-compute-model-attributes%2212:26
dviroelah12:27
dviroelso this one was facing an issue12:27
dviroelwhich was reported here12:27
dviroel#link https://bugs.launchpad.net/watcher/+bug/212058612:27
dviroelan sean-k-mooney noticed that actually, the python-novaclient has its max api version frozen to 2.9612:28
dviroelhttps://bugs.launchpad.net/watcher/+bug/2120586/comments/112:28
amoralejso, that forces us to move to openstacksdk ?12:28
sean-k-mooneyya so the bug itself it trival to fix. and as it a bug that can be doen before or after FF up until RC112:28
dviroelwe could still fix the mentioned bug, and force the api microversion to 2.100 if we want too12:28
sean-k-mooneythe client supprot is harder to adress and need us to swap to the sdk12:29
sean-k-mooneywhich we shoudl do next cycle12:29
dviroelbut that's not aligned with the version supported by the novaclient12:29
dviroelso yes, the proper thing to do is to be limited to 2.96 for now 12:29
dviroeland adress with the openstacksdk in the future12:30
dviroelso I will propose for this cycle, to partially implement the extended attributes12:30
sean-k-mooneyso im +2 on the 2 preceeding patches and was going to merge them after the meeting i fyou want to respine the final patch to limit to 2.96 i woudl be ok with that for this cycle12:30
dviroelwe will miss the scheduler_hints since it was added in 2.10012:30
dviroeli am planning to propose the update today12:31
sean-k-mooney+112:31
dviroelI think that we can already benefit from the pinned_az and the flavor extra_specs 12:31
dviroelfor future strategies developments12:31
dviroelthis will be likely the api 1.612:32
amoralej+1 sounds good12:32
amoralejanything else wrt this one?12:33
sean-k-mooneyyou will need to propose another spec for next cycle 12:33
sean-k-mooneybut i think it may make sense anyway so you can desicbe which stragies will make use of the new info in the same one12:33
amoralejso, https://review.opendev.org/c/openstack/watcher-specs/+/955921 will be cover this cycle?12:34
amoralejor it's part of next cycle work?12:34
dviroeloh yes, there is the addition of the flavor extra_specs, which will be covered12:34
dviroelthis cycle12:34
sean-k-mooneyya so can you update that to note the hint wont be done this cycle as well12:35
sean-k-mooneyamoralej: thanks for reminding me about that i tought we merged the amened ment already12:35
dviroelyes12:35
amoralejok, moving on to next ...12:36
dviroelthat's it then, I will be working on the updates today - thanks for reviewing12:36
amoralej#topic (Flamingo features) Eventlet Removal [MERGED]12:37
amoralejso, i assume you are done with the eventlet removal for this cycle?12:37
dviroelyes, the proposed patches are all merged12:37
dviroelfor this cycle12:37
dviroelnext thing is to work in the applier ones12:37
amoralej#info all the expected work for eventlet removal in flamingo is merged!12:38
sean-k-mooneydviroel: one think i might ask you do do is to write up a short summary of the work so far for the cycle highlights12:38
dviroelsean-k-mooney: sure I can do that12:38
sean-k-mooneygreat there is no rush but we normally strat those after FF12:39
amoralej#topic Aetos datasource [MERGED]12:40
amoralejthat's also done, right?12:40
amoralejany pending task?12:40
amoralej#info new datasource aetos has been merged12:40
dviroel++12:40
sean-k-mooneyyep there is still one followup12:41
sean-k-mooneyto add the aetos job toe the tempest plugin12:41
sean-k-mooneybut the feature itself is complete and that is not bound by the FF deadline12:41
amoralejdo we expect to get this during this cycle?12:41
sean-k-mooneyits a trivial patch so yes12:41
sean-k-mooneyits already running on master for watcher we just need to add it to check in the tempest plugin12:42
amoralejis there support to install and run aetos in devstack ?12:42
sean-k-mooneyyes12:42
dviroelthis one https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/957556 ?12:42
amoralejah, cool12:42
sean-k-mooneyyep thats the one oh its merged12:42
amoralejI totally missed that, sorry12:42
sean-k-mooneythen cool there is nothign mroe for that this cycle then12:43
sean-k-mooneyat the ptg we need to plan a session with telemetry to discuss the future of the promethus direct plugin12:43
amoralejso that was it wrt flamingo features12:43
dviroel+112:43
amoralej#topic Bug Triage12:44
amoralejthere are a couple of new bugs12:44
amoralej#link https://bugs.launchpad.net/watcher/+bug/212066612:44
amoraleji created that, summary, when adding a new volume from notification, it complains that one field is missing12:45
amoralejfield is `multiattach`12:45
amoralejthe volume is added to the model, so the strategies works correctly (as multiattach is currently not used)12:46
amoralejbut an error is displayed12:46
amoralejI couldn't check if the missing field in the notification is expected on cinder notifications, or it may be a bug in cinder12:46
dviroelwe would need to check the documentation in cinder, to check where is the error12:46
sean-k-mooneymultiattach is not a filed12:47
sean-k-mooneyso there are propeties or cablaities on voluem and that is a dict of string to string effectivly12:47
sean-k-mooneyand multi attaahc i belive is one of hte ones that can be set on teh volume type12:48
sean-k-mooneyso we shoudl not assume its there as 99% of voluems wont have it set12:48
amoralejaccording to api-ref on volumes https://bugs.launchpad.net/watcher/+bug/212066612:48
amoralejhttps://docs.openstack.org/api-ref/block-storage/v3/index.html#list-accessible-volumes-with-details12:48
amoralejmultiattach is a mandatory boolean parameter on volumes12:48
amoralejfield i meant12:49
amoralej"If true, this volume can attach to more than one instance."12:49
sean-k-mooneyis it though or is it somethimes null12:49
amoralejactually, to be clear, on the volume created that i got that error, the api reported it as false12:49
sean-k-mooneyhum ok this may have chagned in v312:50
amoralejmultiattach means it's multiattach capable, not that it's actually mutliattched12:50
sean-k-mooneyi think in v2 it was not mandaroy but if its alwasy set now we shoudl handel that12:50
jgilaberbut the bug comes from a notification, I did not see any documentation of the cinder notification format12:51
amoralejso, we may need to call the cinder api during notification handling if we don't get it in the notification12:51
sean-k-mooneythe reason im doubting it is i normally interact with cidner via horizon and it does not show it but that does nto eman it not returned12:51
amoralejyes, i couldn't find it also, although i probably didn't research enough12:52
sean-k-mooneywell the noticaiotn have nothitng to do with the rest apis12:52
amoralejyeah12:52
sean-k-mooneythe files fo the noticiaton do not use the same obejct as the rest apis so while it may be in the api it migh tnot be in the notifcaion12:52
sean-k-mooneyidealy we woudl not call the cinder api in respocne to a notificiton12:52
amoralejyep, but i don't see any other way12:53
sean-k-mooneywe woudl heal any missing information then ext time the perodic runs12:53
amoralejyes, it does12:53
amoraleji fogot that, that's correct, the information is added in the periodic run12:53
amoralejperiodic collection12:53
sean-k-mooneyhttps://paste.opendev.org/show/bxLFmfBTHzTrjzlel9TA/12:54
amoralejif we accept that, we may only want to manage that exception and replace that error by a more friendly warning message12:54
sean-k-mooneyso ya its defintly there in the api responce so we jsut need to be more tolernet to that internally 12:54
sean-k-mooneywell we coudl also ask cidner to include it in the notificaiotn object going forward12:55
amoralejmy doubt is, what if at some point we rely on that field? should we manage the lack of it in the strategy? do some default when handling the notification?12:55
amoralejto avoid errors until next periodic collection12:55
sean-k-mooneyso you cant actully migrate multi attach voluems if they are in use12:56
dviroelif is a mandatory attribute for a strategy, we may need to 1) user cinder api to get the missing information from the notification OR 2) call the collection before running the strategy12:56
sean-k-mooneyso it is somethign we shoudl be aware of12:56
sean-k-mooneyi think if its not present we woudl have to skip the volume12:56
amoralejthat may be a good option12:56
sean-k-mooneysince we are tryign to avoid doing api calls during the evacluation fo the audits12:56
sean-k-mooneyagain this is proably somethign we should raise with the cidner team as a bug/mini feature and see if we or they can implemtne it in the notificaion object next cycle12:57
amoralejin any case, if we call migrate on multiattached volumes, cinder will reject and it would fail, but better to skip12:57
amoralejlet's keep discussing on the bug12:58
sean-k-mooneycidner or nova depending12:58
amoraleji'd like to present next bug, which is related12:58
amoralej#info https://bugs.launchpad.net/watcher/+bug/212114712:58
sean-k-mooneyyou can migrate a multi atach volume if its only attached to 1 vm12:58
amoralejright, actually, what we should check is the attachments, then, not the capability12:58
amoralejthis bug was reported by jgilaber and it's related to an error because of optional field is not found12:59
sean-k-mooneynot quite it can have more then one attachmetn for other reasons.12:59
sean-k-mooneymainly bugs but it can happen12:59
sean-k-mooneybut ya going back to the next bug12:59
jgilaberI found that this morning, when building the xml representation of the storage model13:00
sean-k-mooneythis again seam potenally vallid13:00
jgilaberwatcher assumes all pools have a 'total_volumes' capability, but pretty much only the lvm driver reports it13:00
sean-k-mooneyso we noted before that we cant rely on backend having pools13:00
jgilaberwhile building the model it's fine because it catches the error  https://github.com/openstack/watcher/blob/90f0c2264c4243b4bfa493e4aa371c5315ce163c/watcher/decision_engine/model/collector/cinder.py#L25013:00
sean-k-mooneyand i guess when they do we cant rely on the tootal being reported13:01
jgilaberbut when building the xml it does not https://github.com/openstack/watcher/blob/90f0c2264c4243b4bfa493e4aa371c5315ce163c/watcher/decision_engine/model/element/base.py#L5613:01
sean-k-mooneybut i wonder are there other api requests we can make to list the voluem in a pool/backend and count them ourselves13:01
jgilaberI think the solution here is to catch the error in the element/base.py13:02
amoralejcurrently, i think that field is not used anywhere, so i think we should make watcher being able to smoothly manage if there is no value rpeported13:02
jgilaberyes, AFAIK that field is only printed in the xml13:02
sean-k-mooneyamoralej: is it not used for the volume blancing stragey 13:02
sean-k-mooneyi guess that might just use teh size rather then count13:03
jgilaberfrom a quick glance I don't think it's used13:03
amoralejyes13:03
sean-k-mooneyi guess for now we can just make this skip if the value is not aviable13:03
jgilaberI don't see any reference in https://github.com/openstack/watcher/blob/master/watcher/decision_engine/strategy/strategies/storage_capacity_balance.py13:03
amoralejit uses total_capacity_gb13:04
sean-k-mooneyack 13:04
sean-k-mooneywhat is as_xml_element used for13:04
amoralejhttps://github.com/openstack/watcher/blob/90f0c2264c4243b4bfa493e4aa371c5315ce163c/watcher/decision_engine/model/notification/cinder.py#L4213:04
amoralejwhich comes from a different field13:05
amoralejsorry, we are running out of time13:05
jgilabersean-k-mooney, to generate an xml string of the model e.g https://github.com/openstack/watcher/blob/90f0c2264c4243b4bfa493e4aa371c5315ce163c/watcher/decision_engine/model/model_root.py#L22513:05
sean-k-mooneyya i was just looking at that13:05
sean-k-mooneyi guess we are storign the model as xml blobs in the db then13:05
sean-k-mooneyits use dfor to_string13:06
sean-k-mooneyas well13:06
sean-k-mooneyi think this might mainly be used for debuging13:06
sean-k-mooneybut we shoudl look into thi smore seperatly13:06
amoralejI'd be in favor of following the approach from jgilaber of managing the exception or even remove it from the model, it can be risky to have fields that are not implemented in some backends13:06
jgilaberyes, all uses I've seen are for logging13:07
sean-k-mooneyso a more python approch woudl eb to implemnet proper __repr__ and __str__ functions13:07
sean-k-mooneyso if we confirm its only for logging 13:07
sean-k-mooneyi woudl be incldine to rip out all the xml logic13:07
sean-k-mooneyand jsut make the pyton object pretty printable instead13:08
sean-k-mooneyif we need it for storing the objects in the db then we can revisti that seperatly13:08
sean-k-mooneyamoralej: and yes i agree wew shoudl avoid havign backend depent feild13:09
amoralejok, i'm closing the mtg, we can keep discussing in the bug or in irc later13:09
sean-k-mooneyif its part fo the model adn not directly retured int eh current api query we shoudl either consdier removing it and or calulating it another way13:09
sean-k-mooney+113:10
amoralejyeah, that wfm13:10
amoralejthanks all for joining13:10
amoralejand thanks dviroel for volunteering to chair next one13:10
amoralejand sorry for not managing time properly! :)13:11
amoralej#endmeeting13:11
opendevmeetMeeting ended Thu Aug 21 13:11:31 2025 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)13:11
opendevmeetMinutes:        https://meetings.opendev.org/meetings/watcher_irc_meeting__august_21st__2025/2025/watcher_irc_meeting__august_21st__2025.2025-08-21-12.05.html13:11
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/watcher_irc_meeting__august_21st__2025/2025/watcher_irc_meeting__august_21st__2025.2025-08-21-12.05.txt13:11
opendevmeetLog:            https://meetings.opendev.org/meetings/watcher_irc_meeting__august_21st__2025/2025/watcher_irc_meeting__august_21st__2025.2025-08-21-12.05.log.html13:11
sean-k-mooneydviroel: amoralej which seres did we decied was getting api microverion 1.5 vs 1.613:35
sean-k-mooneyis the skip seriese first ?13:35
dviroelskip can get the 1.5 first13:36
sean-k-mooneyack ill review them in that order then13:36
dviroel++ tks 13:36
sean-k-mooneythere might be a mege conflict between the two so ill prefer merging the skip serise if your fine to rebase your changes 13:36
dviroelsure, I am alredy expecting that, no problem at all13:37
dviroelthere will be a minor conflict in the api yes13:37
amoralejthanks dviroel++13:39
sean-k-mooneyamoralej: im leaving comment for the issue that i see in the skip serise but i dont see anythng so far that cant be adressed in fa follow up patche later14:06
amoralejgreat14:07
sean-k-mooneywhen your back form pto it woudl be nice to adress them assuming noone else has time 14:07
sean-k-mooneythanks for breakign up the patche as we requeted before its making it much simpler to review them14:08
amoralejyeah, it makes much more sense14:08
sean-k-mooneyso this is existing tech debty whichim not gong to ask you to fix in this seres14:11
sean-k-mooneybut for future refence when bumping OVOs like https://review.opendev.org/c/openstack/watcher/+/956705/2/watcher/notifications/action.py14:11
opendevreviewMerged openstack/watcher master: Add options to disable migration in host maintenance  https://review.opendev.org/c/openstack/watcher/+/95253814:11
sean-k-mooneyyou should implement a convertion fucntion to backlevel the object to the old version14:11
sean-k-mooneywatcher does nto do that ata all today14:11
sean-k-mooneybut it a fundemetnal part of OVOs design14:12
sean-k-mooneythis is what that looks like https://github.com/openstack/nova/blob/master/nova/objects/instance.py#L234-L25014:12
amoralejI was about to ask you for an example :)14:13
amoralejbut that's only required in objects or also in notifications?14:13
sean-k-mooneythis is a better one https://github.com/openstack/nova/blob/master/nova/objects/image_meta.py#L20814:13
sean-k-mooneyamoralej: in both14:13
sean-k-mooneyor at least it should be 14:13
sean-k-mooneythe inten of that is if i have code that can only handle teh v1.0 version i can ask the object to downgrade to that14:14
sean-k-mooneybut again watcher does not use that today at all14:14
sean-k-mooneyeven though this si the primary reason oslo.versionedobject were created14:15
sean-k-mooneyim undecied if ro when we shoudl adress that tech debt as we have higher proirty thing to fix14:15
amoralejack14:15
sean-k-mooneythe only reaons this does nto break us today si we do not supprot runnnign diffenet version fo applier decsion enghine and api14:16
sean-k-mooneyfor nova, neutron and cidner it way more impoant as teh compute agent cinder voluem adn neutron agent on the comptue can be an older release teh the contoler services14:16
opendevreviewMerged openstack/watcher master: Add patch call validation based on allowed_attrs  https://review.opendev.org/c/openstack/watcher/+/95599914:24
opendevreviewMerged openstack/watcher master: Add parameters to force failures in nop action  https://review.opendev.org/c/openstack/watcher/+/95581314:32
opendevreviewMerged openstack/watcher master: Add `status_message` column to Actions, Audits and ActionPlans tables  https://review.opendev.org/c/openstack/watcher/+/95474514:50
opendevreviewAlfredo Moralejo proposed openstack/watcher-dashboard master: [WIP] Add option to SKIP Actions  https://review.opendev.org/c/openstack/watcher-dashboard/+/95820914:51
dviroelamoralej: i have 2 questions on https://review.opendev.org/c/openstack/watcher/+/954746/8#message-b9f75018a2bc2f588acbff42a871e4cc00f8a6aa - I was a +2 before that, but It came to my mind while reviewing other change14:54
amoralejchecking14:54
dviroelamoralej: revert and abort could be skipped I think. At the same time, the action implementation could check if something needs to be reverted or not, and return properly.14:57
dviroelso I guess that is more an improvement that you could provide later14:57
amoralejI assumed that do_revert and do_abort would never been called if never go throught the execute14:57
amoralejbut I'm probably wrong14:58
dviroelthe workflow doesn't know, because you return True in the execute14:59
opendevreviewMerged openstack/watcher master: Add `status_message` to objects and notifications  https://review.opendev.org/c/openstack/watcher/+/95670514:59
dviroeliiuc14:59
amoralejso, taskflow may call abort or revert, then make sense to add the condition there too, in any case it will not hurt15:00
dviroelif sean-k-mooney agree with a follow up for that too, we should not consider in this patch, you can propose as a improvement later15:03
dviroelamoralej: ^15:03
sean-k-mooneyso if the action is in the skiped state15:09
sean-k-mooneythen revert adn aboort should be no ops15:09
sean-k-mooneyregardess of what the action actully is15:09
dviroelyes15:09
opendevreviewRonelle Landy proposed openstack/watcher master: Update Overload standard deviation doc  https://review.opendev.org/c/openstack/watcher/+/95406715:09
sean-k-mooneyso i guess i need to look at it agin but we likely shoudl enforce that before compelting this feature15:11
sean-k-mooneyit should be relitivly simple ot extend the basse action class with a method to do this and then call that in abort/revert to skip if needed15:12
dviroelsure, at your time15:12
sean-k-mooneyamoralej: gneral comment on https://review.opendev.org/c/openstack/watcher/+/955753/5/watcher/api/controllers/v1/action.py#10615:12
sean-k-mooneyyou shoudl never use hassatr or gettattr in production code if there is any way to aovid it15:12
sean-k-mooneytaht goes double for setattr15:13
sean-k-mooneythis is proably accpable for now but i consider all usage of hassatr, gettr and setatr ot be techinial debt that shoudl be cleanded up eventually15:15
*** dviroel_ is now known as dviroel20:55
*** gmaan_ is now known as gmaan20:55

Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!