| *** wuchunyang has joined #zuul | 01:26 | |
| *** wuchunyang has quit IRC | 01:29 | |
| *** swest has quit IRC | 01:51 | |
| *** swest has joined #zuul | 02:06 | |
| *** bhavikdbavishi has joined #zuul | 02:46 | |
| *** bhavikdbavishi1 has joined #zuul | 02:52 | |
| *** bhavikdbavishi has quit IRC | 02:55 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 02:55 | |
| *** wuchunyang has joined #zuul | 04:07 | |
| *** bhagyashris|afk is now known as bhagyashris | 04:24 | |
| *** sgw1 has quit IRC | 04:28 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #zuul | 04:33 | |
| *** raukadah is now known as chandankumar | 04:42 | |
| *** ysandeep|away is now known as ysandeep | 04:43 | |
| *** wuchunyang has joined #zuul | 04:44 | |
| *** wuchunyang has quit IRC | 05:01 | |
| *** saneax has joined #zuul | 05:36 | |
| *** marios has joined #zuul | 05:58 | |
| *** sshnaidm|off is now known as sshnaidm|ruck | 06:17 | |
| *** ysandeep is now known as ysandeep|brb | 06:21 | |
| *** felixedel has joined #zuul | 06:26 | |
| *** ysandeep|brb is now known as ysandeep | 06:34 | |
| *** bhavikdbavishi has quit IRC | 06:37 | |
| zbr | avass: corvus clarkb: please help me finish the ANSI enablement, https://review.opendev.org/#/c/716251/ | 07:05 |
|---|---|---|
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul master: Enable ANSI rendering on stdout/stderr https://review.opendev.org/716251 | 07:12 |
| *** saneax has quit IRC | 07:16 | |
| *** felixedel has quit IRC | 07:18 | |
| *** saneax has joined #zuul | 07:18 | |
| *** bhavikdbavishi has joined #zuul | 07:20 | |
| *** saneax has quit IRC | 07:22 | |
| *** tosky has joined #zuul | 07:25 | |
| avass | zbr: I might be able to take a look at it in a moment. but javascript isn't really my domain :) | 07:30 |
| zbr | mine neither, doing it only for the sake of zuul. | 07:31 |
| *** jcapitao has joined #zuul | 07:35 | |
| *** nils has joined #zuul | 08:06 | |
| *** bhavikdbavishi1 has joined #zuul | 08:09 | |
| *** bhavikdbavishi has quit IRC | 08:10 | |
| *** holser has joined #zuul | 08:11 | |
| *** bhavikdbavishi has joined #zuul | 08:12 | |
| *** bhavikdbavishi1 has quit IRC | 08:14 | |
| *** marios has quit IRC | 08:21 | |
| *** ysandeep is now known as ysandeep|lunch | 08:23 | |
| *** bhavikdbavishi has quit IRC | 08:35 | |
| *** bhavikdbavishi has joined #zuul | 08:38 | |
| *** persia has quit IRC | 08:44 | |
| *** persia has joined #zuul | 08:45 | |
| *** bhagyashris is now known as bhagyashris|afk | 08:50 | |
| *** hashar has joined #zuul | 08:53 | |
| *** ysandeep|lunch is now known as ysandeep | 08:56 | |
| *** marios has joined #zuul | 09:16 | |
| *** bhagyashris|afk is now known as bhagyashris | 09:19 | |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul master: Enable ANSI rendering via react-ansi https://review.opendev.org/739444 | 09:42 |
| zbr | mhu: also check ^, different approach. | 09:49 |
| mhu | zbr, I'll wait for the preview to be generated, but my general concern about having it disabled by default remains | 09:58 |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul master: Replace cookie use with localStorage https://review.opendev.org/739454 | 10:12 |
| *** rfolco has joined #zuul | 10:37 | |
| *** jcapitao is now known as jcapitao_lunch | 10:57 | |
| openstackgerrit | yatin proposed zuul/zuul-jobs master: [DNM] Test task in a role https://review.opendev.org/739463 | 11:02 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Scheduler's pause/resume functionality https://review.opendev.org/709735 | 11:43 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Separate connection registries in tests https://review.opendev.org/712958 | 11:43 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Prepare Zookeeper for scale-out scheduler https://review.opendev.org/717269 | 11:44 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Mandatory Zookeeper connection for ZuulWeb in tests https://review.opendev.org/721254 | 11:44 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect merger to Zookeeper https://review.opendev.org/716221 | 11:44 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect fingergw to Zookeeper https://review.opendev.org/716875 | 11:44 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect executor to Zookeeper https://review.opendev.org/716262 | 11:44 |
| *** olaph has joined #zuul | 11:48 | |
| *** bhavikdbavishi has quit IRC | 11:52 | |
| *** pabelanger has left #zuul | 12:02 | |
| *** rlandy has joined #zuul | 12:05 | |
| *** jcapitao_lunch is now known as jcapitao | 12:08 | |
| *** bhavikdbavishi has joined #zuul | 12:18 | |
| *** ysandeep is now known as ysandeep|afk | 12:38 | |
| *** rlandy is now known as rlandy|ruck | 12:45 | |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul master: WIP: Add user preferences dialog https://review.opendev.org/739482 | 12:47 |
| *** ysandeep|afk is now known as ysandeep | 12:56 | |
| *** Goneri has joined #zuul | 12:57 | |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul master: Replace cookie use with localStorage https://review.opendev.org/739454 | 13:17 |
| *** rlandy|ruck is now known as rlandy|ruck|mtg | 13:30 | |
| *** bhavikdbavishi has quit IRC | 13:37 | |
| *** nils has quit IRC | 13:43 | |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Introduce Patternfly 4 https://review.opendev.org/736225 | 13:46 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: PF4: Update "fetching info ..." and refresh animation https://review.opendev.org/738010 | 13:46 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: PF4: Update buildset result page (new layout and styling) https://review.opendev.org/738011 | 13:46 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: PF4: Add new Zuul logo with text https://review.opendev.org/738033 | 13:46 |
| *** felixedel has joined #zuul | 13:46 | |
| felixedel | zuul-maint: Kindly asking for a review on https://review.opendev.org/#/c/736225/ I would really like to get that change in first, so any further web UI updates could utilize Patternfly 4 | 13:48 |
| zbr | was someone working to remove venv req from tox jobs? | 14:00 |
| zbr | felixedel: how close to ready is the pf4 change? is see you keep updating it. | 14:03 |
| felixedel | zbr: The initial PF4 patch is ready for review from my side. I just noticed a missing margin value on the change page today while working on a different change and added it directly to that patch :) | 14:07 |
| zbr | i will try to look at it and see if there are any serious issues, minor stuff can be fixed in followups, IMHO | 14:08 |
| felixedel | Thanks | 14:08 |
| zbr | somehow i see quite hard to merge any changes to UI, i keep hiting blockers, like outdated react, now outdated patternfly, or just hard to get cvorum as almost always I found someone that prefers the "old behavior". | 14:10 |
| felixedel | With that we would at least solve the outdated patternfly issue ;-) And as I feel similar, I really would like to get that change in before I continue further with the PF4 updates as this might result in a lot of rebasing in case someone has complains about this change. | 14:12 |
| fungi | it's not "old behavior" until it's replaced, until then it's "current behavior" ;) | 14:13 |
| zbr | fungi: yeah. hopefully, adding experimental options would make these changes easier to go in. | 14:14 |
| tobiash | zbr: I think your local storage change is a good start to make this possible without having issues because of the cookies | 14:15 |
| zbr | tobiash: that is why I opened it, not because I was personally concerned about GDPR | 14:15 |
| avass | zbr: venv was delayed until july 16 to give ansible some time: http://lists.zuul-ci.org/pipermail/zuul-discuss/2020-June/001264.html | 14:28 |
| *** rlandy|ruck|mtg is now known as rlandy|ruck | 14:29 | |
| *** felixedel has quit IRC | 14:30 | |
| zbr | avass: ok. meanwhile i will manually override it in order to avoid some failures caused by it. | 14:30 |
| zbr | fungi: avass tobiash: please help me kill the cookie monster, https://review.opendev.org/#/c/739454/ -- fixed last comments, seems to work very nice. | 14:31 |
| tobiash | zbr: I already looked at it and wanted to have a look at the preview before voting | 14:32 |
| zbr | sure, please test. works nice in preview,... until someone would reply with "it does not work my my custom build browser in magic mode" ;) | 14:33 |
| zbr | localStorage can be used to keep a lot of data, probably is also much faster than cookies, which often disable some caching | 14:35 |
| zbr | a reverse proxy cannot usually cache requests coming with different cookies, as it would expect possible different server responses based on them | 14:36 |
| corvus | i agree that localstorage is probably a better fit (even though i also don't believe there is a gdpr problem); and the patch works for me in preview | 14:37 |
| corvus | mhu: do you want to take another look or should we +w 739454? | 14:38 |
| avass | corvus: oh since you're here. you might want to release the hold on 158.69.68.12 if you're done with it :) | 14:38 |
| zbr | any legal discussion would only guarantee pain, so better to avoid that debate. | 14:38 |
| corvus | zbr: i don't think it would guarantee pain, i think it would provide clarity | 14:38 |
| zbr | corvus: you are welcomed to lead that quest if you want, i have no spare energy. | 14:40 |
| mhu | corvus, I didn't test the suggestion I made, if you give me a few minutes I can deploy and make sure it works as intended | 14:40 |
| corvus | i don't think it's necessary; there seems to be widespread consensus that our existing and future use of cookies are not subject to gdpr issues | 14:40 |
| corvus | mhu: can you test it with the site preview? | 14:40 |
| mhu | corvus, I'm not sure, when I tried to open tabs from the site preview I got a "file not found" | 14:41 |
| corvus | mhu: yeah, you can't deep link, but maybe you can just open another tab from the base url? | 14:42 |
| mhu | just a moment, let me fetch the change, spawn the dev build and I'll test it out | 14:42 |
| corvus | eg https://ddd63cdd8cfb9f3fb3ba-183a764968d603a8e1bec46158333aa8.ssl.cf2.rackcdn.com/739454/2/check/zuul-build-dashboard-opendev/814664b/npm/html/ | 14:43 |
| corvus | mhu: i opened another tab and the setting did not persist | 14:43 |
| mhu | corvus, well there you have it then :/ | 14:43 |
| corvus | could that be a quirk of the preview site? | 14:44 |
| mhu | I had a hunch I might have fumbled it, sorry zbr | 14:44 |
| zbr | corvus: can you send an email and CC me to foundation lawyer? https://gdpr.eu/cookies/ has written black on white: MUST Receive users' consent before you use any cookies except strictly necessary cookies. | 14:44 |
| corvus | mhu, zbr: oh wait, i was wrong; i think it is working | 14:45 |
| mhu | corvus, it seems to be working on my dev build too | 14:45 |
| zbr | and TZ cookie is clearly not a strictly necessary, it counts as a "preference" one. | 14:45 |
| mhu | so I'm good with +w | 14:45 |
| corvus | zbr: this is what a lawyer would call a "moot" point and therefore not actionable. | 14:45 |
| corvus | zbr: so no, i won't write your email for you. but i will land your change. | 14:46 |
| corvus | zbr: i think it's a good change, thanks :) | 14:47 |
| mhu | if we're having a web UI review party, I'd like to get some opinions on https://review.opendev.org/#/c/737290/ (i18n) | 14:48 |
| zbr | thanks. be extra careful with my js/react changes, i google a lot for each line i write. | 14:48 |
| corvus | zbr: the life of a modern programmer :) | 14:48 |
| corvus | mhu: ooh i'm digging the _ | 14:49 |
| mhu | It's a big change to maintain so I'd like to have it merged sooner than later - especially if the patternfly 4 lands | 14:49 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Correctly handle skipped jobs on retries https://review.opendev.org/739511 | 14:49 |
| zbr | corvus: i am curious what you think about https://review.opendev.org/#/c/739482/ -- not much to see in preview (empty dialog) but i am curious about the idea/concept. | 14:49 |
| corvus | mhu: i was also thinking it should be possible to make a component right? like "<t>Translated text</t>" ? | 14:50 |
| zbr | i need to know if I start moving config items to it. | 14:50 |
| corvus | mhu: is that at all useful? | 14:50 |
| mhu | corvus, it wouldn't be much different from the original <Translate > that comes from the react lib then? | 14:51 |
| mhu | except the translated text has to be passed as a prop | 14:52 |
| mordred | corvus: I think we should change the official zuul term for config project to "Environnement d'exécution sûr" | 14:52 |
| zbr | nobody implemented a _() function like on gettext for react? that was pretty neat. | 14:52 |
| corvus | zbr: i like the concept (a lot actually); i think mhu suggested that we put it behind the user login change? login won't be required, but it seems like that's further along and they might conflict a bit. | 14:53 |
| * mordred likes the preferences dialog concept a lot too | 14:54 | |
| corvus | mhu: well, my biggest concern was the strings being constants. i think the most important thing is "just have the strings inline"; then whether you use "<Translate>", "<t>" or "{_()}" is gravy | 14:54 |
| zbr | i doubt it would not be easy to migrate the preferences to something compatible when user login lands. also lots of admins will not want to allow logins anyway | 14:54 |
| mhu | mordred: shall I add you to the unofficial list of french translators? :) | 15:00 |
| corvus | i think mordred was suggesting we adopt the french term universally :) | 15:01 |
| zbr | while I do have a long experience in i18n, it does not cover JS world. still same issues apply everywhere. | 15:01 |
| mordred | mhu: oui. le chat mange! | 15:01 |
| mhu | mordred, le chat mange dans un contexte d'exécution sûr? | 15:01 |
| zbr | I do not see any utility to extract and update language strings | 15:02 |
| mordred | mhu: bien sûr | 15:02 |
| zbr | french will be fun, "pass build" or "pas build" ;) | 15:03 |
| mhu | corvus, most strings I came upon were constants, but there was the occasional templated string (page titles mostly), and singles/plurals handling (which was non-existent before so it's an improvement) | 15:04 |
| mhu | zbr, actually some translations (or non-translations) to french were head scratchers, I'll be counting on other french users to give feedback on them | 15:04 |
| fungi | https://messageformat.github.io/Jed/ | 15:05 |
| corvus | mhu: sorry i meant referring to the strings by symbolic constants in the code (ie "<Translate text="SYMBOLIC_CONSTANT">" | 15:05 |
| mhu | corvus, oh right | 15:06 |
| zbr | corvus: tbh, I would be against using constants. Every project that used that become a PITA to maintain as it required two step searches | 15:06 |
| *** hashar has quit IRC | 15:06 | |
| mhu | it's just a few of them IIRC, templates and a very long string | 15:07 |
| corvus | zbr: yes, that was my position too, which is why i suggested mhu look for alternatives, which he did, and the new patch is much improved. | 15:07 |
| mordred | mhu: just to make things potentially complex for your french translation - we've discussed adding a 3rd type of project, a "tenant config project" which would be like a tenant-specific config project but which does not allow executing code on executors. I don't know if that would make the "safe execution context" and "unsafe execution context" translations become awkward | 15:07 |
| corvus | fungi, mhu: jed looks interesting | 15:07 |
| * corvus breakfasts | 15:07 | |
| zbr | this is what i would have used: https://www.npmjs.com/package/react-gettext -- i have no idea how good is this impementation, but I know that gettext does i18n very well. | 15:08 |
| mhu | it has to play well with redux as well | 15:09 |
| zbr | which is not easy, only google about plural strings and you may need a vacation to recover | 15:09 |
| mhu | the first lib I used crashed completely unrelated parts of the UI | 15:09 |
| mordred | \o/ | 15:09 |
| zbr | maybe switching to pf4 fist would be a good idea | 15:10 |
| mhu | As much as it pains me just thinking of what rebasing will be like, I agree :) | 15:10 |
| zbr | we do not want to endup not being able to upgrade pf due to an outdated i18n library | 15:10 |
| zbr | corvus: mordred: any of you against moving the TZ selector under the config dialog? or you think is key to be always visible on screen? | 15:18 |
| zbr | i wonder if there is a way to register preferences from various places in the app, and render them in a dialog without known about each of them. | 15:20 |
| zbr | so we would not have to update the preferences dialog when adding a new user configurable option. | 15:20 |
| corvus | zbr: we probably need to at least display it on screen all the time | 15:20 |
| zbr | i was afraid you may say so. i need one option to add to my dialog, i doubt I can merge it is as is, with zero preferences in it. | 15:21 |
| fungi | it makes sense for the ui to always indicate what tz any times it's displaying are relative to (for sake of clarity/specificity), and at that point it may as well also be a clickable ui element for switching on the fly | 15:22 |
| *** ysandeep is now known as ysandeep|away | 15:22 | |
| zbr | how many computers or websites do display the current TZ by default? | 15:23 |
| zbr | i will probably start by moving the auto-reload button from status page, that one should be less controversial | 15:23 |
| fungi | as someone who always wants to see times in utc, i assume all times are in utc unless otherwise specified | 15:24 |
| *** sshnaidm|ruck is now known as sshnaidm|mtg | 15:24 | |
| fungi | but if an interface doesn't indicate, then that's frustratingly vague | 15:24 |
| fungi | most services also aren't displaying logs and events intended for correlation | 15:24 |
| zbr | fungi: same with me. you cannot imagine what kind of endless debates I had in the past regarding setting servers to TZ. | 15:24 |
| zbr | many said i was crazy, until they had to move the VM to another region :D | 15:25 |
| fungi | what, you don't want to know whether it's nighttime where your virtual machine is located: ;) | 15:26 |
| *** sgw1 has joined #zuul | 15:27 | |
| mordred | zbr: anyone who argued against you about setting server timezones to UTC was wrong ;) | 15:33 |
| *** hamalq has joined #zuul | 15:36 | |
| *** holser has quit IRC | 15:39 | |
| *** holser has joined #zuul | 15:39 | |
| openstackgerrit | Albin Vass proposed zuul/zuul master: Add swedish locale https://review.opendev.org/739524 | 15:52 |
| avass | :) | 15:52 |
| *** marios is now known as marios|out | 15:52 | |
| *** hashar has joined #zuul | 15:52 | |
| mhu | avass, any trouble with adding the locale? | 15:53 |
| avass | mhu: well, mostly the swedish language and how we use english words for everything ;) | 15:54 |
| avass | but no not really, I think it was pretty straight forward | 15:54 |
| *** marios|out has quit IRC | 15:54 | |
| mhu | avass, my swedish is very rusty, is it normal that plurals are the same as singles, for example events_* ? | 15:55 |
| avass | it depends on the word really | 15:56 |
| avass | I might need to get input from someone else on this first since some things are still a bit strange. like 'change' would technically be 'förändring' but no one says that and uses change instead. | 15:56 |
| mhu | I had the exact same questioning | 15:57 |
| fungi | it's the case in many, many languages. english terms have been used so extensively in technical contexts for so long that users find it strange seeing them localized, i'm told | 15:58 |
| mhu | true | 15:59 |
| tobiash | definitely, e.g. I cannot work with a german horizon, I always switch back to english | 15:59 |
| avass | absolutely, no one would understand what I meant if I used the swedish words for a lot of things | 16:00 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Scheduler's pause/resume functionality https://review.opendev.org/709735 | 16:12 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Separate connection registries in tests https://review.opendev.org/712958 | 16:12 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Prepare Zookeeper for scale-out scheduler https://review.opendev.org/717269 | 16:12 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Mandatory Zookeeper connection for ZuulWeb in tests https://review.opendev.org/721254 | 16:12 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect merger to Zookeeper https://review.opendev.org/716221 | 16:12 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect fingergw to Zookeeper https://review.opendev.org/716875 | 16:12 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect executor to Zookeeper https://review.opendev.org/716262 | 16:12 |
| *** jcapitao has quit IRC | 16:16 | |
| *** bhavikdbavishi has joined #zuul | 16:18 | |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul master: WIP: Add user preferences dialog https://review.opendev.org/739482 | 16:19 |
| *** sgw1 has quit IRC | 16:21 | |
| *** sgw1 has joined #zuul | 16:23 | |
| avass | zuul-jobs-maint: we should merge: https://review.opendev.org/#/c/727158/ by the way :) | 16:25 |
| *** sgw1 has quit IRC | 16:31 | |
| *** bhavikdbavishi1 has joined #zuul | 16:33 | |
| *** sgw1 has joined #zuul | 16:34 | |
| *** bhavikdbavishi has quit IRC | 16:35 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 16:35 | |
| zbr | are these zuul-upload-image errors exppected? https://zuul.opendev.org/t/zuul/builds?job_name=zuul-upload-image&project=zuul/zuul | 17:00 |
| zbr | looks like a very bad success rate for a gate job | 17:00 |
| *** harrymichal has joined #zuul | 17:01 | |
| zbr | dmsimard: do you know any reasons why we should keep the popup on result-label and not to put it as expandable task info? | 17:08 |
| zbr | now the extend section has few fields than the full popup, but the difference is kinda small, would be much easier to have a single display method. | 17:09 |
| zbr | i would even try to change how we display to waste less space on screen, most fields are single line so better displayed in column mode (would reduce vert space to ~1/3 for these) | 17:10 |
| zbr | is it quite easy to figure out what field is small or big. | 17:10 |
| *** hashar has quit IRC | 17:22 | |
| openstackgerrit | Merged zuul/zuul-jobs master: Make persist-iptables more robust https://review.opendev.org/739061 | 17:33 |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul master: Reduce table nesting on build pages https://review.opendev.org/739559 | 17:34 |
| dmsimard | zbr: I don't have a strong opinion on UI/UX | 17:35 |
| openstackgerrit | Sorin Sbarnea (zbr) proposed zuul/zuul master: Reduce table nesting on build pages https://review.opendev.org/739559 | 17:37 |
| zbr | dmsimard: i started a process of simplification of the UI, I think that we have too many nested containers visible in the UI | 17:39 |
| zbr | while react may require them, it does not mean that they all need to be visible to the user. | 17:40 |
| mhu | corvus, regarding your comment on the PatternFly4 change, does that mean it's going to be the next "big" change in the UI and all other patches should be rebased on it if needed? | 18:04 |
| mhu | just trying to get a sort of timeline here | 18:04 |
| corvus | mhu: that wasn't really my intent; i'm not sure if it's that disruptive. it has some conflicts, but not a lot. do you think it is? | 18:05 |
| mhu | corvus, I can attempt a rebase and see if there are any problems, it should always be possible to revert back | 18:06 |
| mhu | it's just that in the case of i18n, both changes impact so many files that I'd expect conflicts | 18:07 |
| corvus | mhu: yeah, that one seems likely. i think felix is pretty close, so that might be reasonable. i'm open to suggestions about how to proceed :) | 18:08 |
| *** y2kenny has joined #zuul | 18:09 | |
| y2kenny | If I get "Executing local code is prohibited", is the module in question completely blocked? or is it just an issue with untrusted project and the playbook would work in a config project? | 18:11 |
| corvus | y2kenny: the second | 18:11 |
| y2kenny | corvus: sweet. thanks! | 18:11 |
| y2kenny | (I was trying to use the xml) module | 18:13 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Update Zuul logo https://review.opendev.org/739568 | 18:16 |
| mhu | corvus, I'd suggest working all together to land PF4 first, then look at i18n - if folks want to have a stab at other libs, and also think about the future translation workflow | 18:17 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: scheduler: Fix event process abide hasUnparsedBranchCache argument https://review.opendev.org/739042 | 18:17 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: Fix branch name and project name for ref-updated create/delete https://review.opendev.org/738320 | 18:17 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: FakeGerritChange: Add Change-Id in commit message https://review.opendev.org/739197 | 18:17 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: WIP: Scheduler: Reconfiguration ref-updated create/delete https://review.opendev.org/739198 | 18:17 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: WIP: Scheduler: Reconfiguration ref-updated oldrev+newrev https://review.opendev.org/739078 | 18:17 |
| corvus | mhu: sounds good. i just did my part to help with pf4 ^ :) | 18:17 |
| mhu | as in: "my change added new strings to translate. Now what?" | 18:17 |
| *** bhavikdbavishi has quit IRC | 18:25 | |
| *** bhavikdbavishi has joined #zuul | 18:26 | |
| *** rlandy|ruck is now known as rlandy|mtg | 18:28 | |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: Fix branch name and project name for ref-updated create/delete https://review.opendev.org/738320 | 18:44 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: FakeGerritChange: Add Change-Id in commit message https://review.opendev.org/739197 | 18:44 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: WIP: Scheduler: Reconfiguration ref-updated create/delete https://review.opendev.org/739198 | 18:44 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: WIP: Scheduler: Reconfiguration ref-updated oldrev+newrev https://review.opendev.org/739078 | 18:44 |
| *** hamalq has quit IRC | 18:49 | |
| *** hamalq has joined #zuul | 18:49 | |
| *** bhavikdbavishi has quit IRC | 18:52 | |
| *** sshnaidm|mtg is now known as sshnaidm|ruck | 19:01 | |
| *** rlandy|mtg is now known as rlandy | 19:17 | |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Update Zuul logo https://review.opendev.org/739568 | 20:21 |
| corvus | tobiash: it seems like we may need one more zuul 3.x release to get the new kazoo once that's released, that way there's a tagged container image with that in it (otherwise, we can't really require zk tls, can we?) | 20:38 |
| tristanC | corvus: do we want to create zuul tag to update the zuul image? | 20:40 |
| corvus | if that's the case, then i think our timeline is: wait for kazoo, upgrade opendev to tls zk, issue 3.19.1 (and nodepool 3.13.1) then start working on 4.x | 20:40 |
| corvus | tristanC: what do you mean? | 20:40 |
| tristanC | corvus: zuul-3.19.0 (source) release doesn't pin the kazoo version, thus installing or updating it after a new kazoo release would be enough | 20:42 |
| tristanC | corvus: would we tag 3.19.1 over the 3.19.0 ref? | 20:42 |
| fungi | that's a good point, it's a packaging revision, unless we want to update the minimum kazoo version in requirements | 20:43 |
| corvus | tristanC: ah gotcha. that's a good point, we could just double tag the same sha | 20:43 |
| corvus | we'll get the image build we want, with minimal other changes | 20:43 |
| corvus | that should mean we don't have to worry about landing big changes | 20:43 |
| corvus | (so we could, for example, merge the pf4 change, which i don't think should be going into zuul v3) | 20:44 |
| tristanC | corvus: but wouldn't that be a release without a note? Perhaps we need a change that force kazoo>2.7.0 requirement min ver | 20:44 |
| corvus | i wonder what reno will do in that case. | 20:46 |
| fungi | that could also be done with a temporary branch... make stable/3.19 from the 3.19.0 tag, merge separate (identical other than their parents) kazoo minimums to stable/3.19 and master, tag that on stable/3.19 as 3.19.1 and then delete the stable/3.19 branch? | 20:46 |
| mordred | corvus, tristanC: I think a min-bump might be a nice thing though - as a way to signal that the newer kazoo is _important_ | 20:47 |
| corvus | i would be okay waiving the 'no release without a note' rule in this case. we're going to have a lot of messaging about it (sending out emails telling people about the upgrade roadmap, etc). but i'm also okay with an explicit new release. | 20:48 |
| mordred | but - I don't think we have to and could also totally go for a double-tag | 20:48 |
| corvus | fungi: you make it sound not very hard. maybe we should do that. | 20:48 |
| corvus | mordred: we're in violent "meh" :) | 20:48 |
| mordred | yeah. | 20:48 |
| mordred | I think I like fungi's thing perhaps | 20:48 |
| corvus | me too | 20:48 |
| fungi | it's a dance openstack used to do often for libary point releases before they just started branching them all in case they needed one | 20:49 |
| corvus | i think we have consensus on that for now. | 20:49 |
| fungi | er, library | 20:49 |
| tristanC | if it's not too much work, then fungi's proposition seems ideal | 20:49 |
| mordred | are we getting a new kazoo release? | 20:49 |
| corvus | 1) kazoo release; 2) branch stable/3.x from 3.19.0; 3) land kazoo version bump with release note; 4) tag 3.19.1; 5) delete stable/3.x | 20:50 |
| corvus | mordred: they seem amenable to doing that this week: https://github.com/python-zk/kazoo/issues/621 | 20:50 |
| fungi | in theory reno should be fine with it, since the 3.19.1 release notes are discoverable from a tag (otherwise reno would be broken for any openstack projects which deleted branches after eol) | 20:50 |
| corvus | or next, depending on the time zone where that comment was written | 20:50 |
| corvus | fungi: yeah; i don't know what reno would do on a double tagged sha | 20:51 |
| corvus | maybe just "3.19.1\n3.19.0\nFeatures..." | 20:51 |
| corvus | but i guess we won't find out :) | 20:51 |
| fungi | we could find out, it's all doable locally without pushing anywhere | 20:52 |
| corvus | yep | 20:52 |
| fungi | but i have no reason to find out | 20:52 |
| fungi | other than morbid curiosity | 20:52 |
| fungi | and i'm morbid enough already | 20:52 |
| mordred | corvus: woot! (re release) | 20:53 |
| *** hashar has joined #zuul | 20:54 | |
| y2kenny | is there anyway to search the log of this irc? (I am trying to find an answer answered here previously) Googling with eavesdrop don't seems to yield all the result | 21:13 |
| mnaser | y2kenny: http://eavesdrop.openstack.org | 21:16 |
| mnaser | "IRC Channel Logs" | 21:16 |
| mnaser | http://eavesdrop.openstack.org/irclogs/%23zuul/ :) | 21:16 |
| y2kenny | mnaser: right... I am trying to find a search function... Google seems to only index some of the pages... | 21:18 |
| mnaser | oh, like a search across eavesdrop.. oh | 21:18 |
| mnaser | y2kenny: have you tried using the google modifiers so i would search... site:eavesdrop.openstack.org inurl:zuul foobar | 21:19 |
| mnaser | and that should search for foobar for everything that's indexed but google but eh | 21:20 |
| mnaser | i don't think we have an eavesdrop search per say | 21:20 |
| y2kenny | I did... and I only find one result with my name which is not a lot... I wonder how google decide what to index... | 21:20 |
| y2kenny | I just ended up wget the logs and grep :P | 21:22 |
| fungi | yeah, better searchable irc logging for opendev has been on my to do list for a very long time, but so many other things always take precedence | 21:25 |
| fungi | i mean, the logging itself is fine (modulo the bot itself needing an overhaul), but the presentation could be improved and some searchable index would be really, really useful | 21:26 |
| fungi | the main reason it hasn't been any higher priority for me is that i keep my client connected and joined to all the channels i care about every minute of the day (excepting server outages and netsplits), so i can locally grep the logs for any channel | 21:27 |
| *** harrymichal has quit IRC | 21:27 | |
| y2kenny | fungi: I have been meaning to get more advance usage with irc myself but freenode has been a bit too easy... :) | 21:28 |
| y2kenny | question about fan out job. If I want to run the same playbook across a range of label/node type or a range of variables, do I have to define a seaparte job for each of those label/variable? Is there some shortcut to spawn multiple job from one job definition? | 21:50 |
| fungi | there is (currently) no job "generator" to create multiple jobs from a matrix of values, though it has come up before as something somebody might want to work on adding | 21:53 |
| y2kenny | fungi: ok thanks. I was wondering if there are some yaml hack / syntactic thing that I can try. | 21:54 |
| y2kenny | I did not know there is pointer in yaml until this year :) | 21:55 |
| fungi | you can use yaml anchors to reuse lists of variables within the same file, but beyond that i'm not aware of any real magic in the yaml spec | 21:55 |
| y2kenny | ok | 21:55 |
| fungi | er, well, reuse any construct really | 21:55 |
| fungi | pointers, yeah that's what i was thinking of | 21:55 |
| tristanC | y2kenny: there is this some additional infos about multiple job definition in this review https://review.opendev.org/#/c/708406/ | 21:56 |
| y2kenny | tristanC: that looks pretty cool | 21:56 |
| y2kenny | switching gear a bit... for nodepool, are there any function within the nodepool framework that allows the driver to do something just before a node is assigned to be used? | 21:58 |
| y2kenny | I see 'launch' which makes the node ready to be use by a job but I don't think there's other thing (I may have missed it) | 21:59 |
| fungi | there used to be a concept of "ready scripts" for nodepool, looking now to see if those still exist | 21:59 |
| y2kenny | ready script.... let me see... | 21:59 |
| *** Goneri has quit IRC | 22:00 | |
| y2kenny | the use case I am trying to handle is that a node that have launch may have gone stale (something have timeout for example) and the node is no longer useable. So I want to do a check just prior to the node getting consumed (or if there's another way to check the freshness of the launched node I am open to that as well.) | 22:00 |
| fungi | it looks like https://review.openstack.org/445567 removed ready scripts (they were a thing before nodepool v3) | 22:02 |
| fungi | you could add a pre phase playbook in your base job which checks the node's freshness. if a pre playbook fails zuul should retry with another node request | 22:03 |
| y2kenny | um... ok | 22:04 |
| fungi | ready scripts weren't really that anyway, they ran after a node was launched before it was marked ready (available) | 22:04 |
| fungi | so wouldn't have caught stale nodes | 22:04 |
| tristanC | y2kenny: perhaps the `max-ready-age` attribute would work? | 22:05 |
| fungi | yeah, if it's strictly age you're concerned with, that's the way to go | 22:05 |
| fungi | if you're worried the node might have died while it was waiting to be used, then not necessarily helpful | 22:05 |
| y2kenny | tristanC: let me take a look at that... | 22:05 |
| fungi | but also if the node is dead and zuul can't connect to it, then it will also retry with a new node request anyway | 22:06 |
| y2kenny | oh the max-ready-age might be useful | 22:06 |
| y2kenny | fungi: that is true but the cobbler driver is doing something that is a bit... funky... I might need to re-think some of the stuff | 22:07 |
| y2kenny | currently it behave kind of like the k8s namespace type | 22:08 |
| y2kenny | it will pretty much always return something success but some of the data has a timeout | 22:08 |
| fungi | oh, ick | 22:09 |
| y2kenny | (I try to let the driver generate a auth token to talk to cobbler and then pass that up to the job for consumption but that token has a limited life time. | 22:09 |
| y2kenny | ) | 22:09 |
| y2kenny | it would be kind of like the k8s driver creating the namespace and create all the certs and such but those certs expires in one hour. | 22:10 |
| fungi | yeah, max age is probably good for that then | 22:11 |
| y2kenny | so max-ready-age will definitely help side step this. | 22:11 |
| y2kenny | but then now I think... what if the job itself take longer than an hour... but if that's the case, even the API I was asking about at the beginning won't help | 22:11 |
| y2kenny | since the token is valid at the beginning of the job. | 22:12 |
| y2kenny | I probably need to rethink where the auth/security should happen | 22:12 |
| fungi | i guess the token can't be used to renew itself | 22:12 |
| y2kenny | it can and should but I still run into problem with the token expiring. And I also can't rule out the job being longer than the renew period. | 22:14 |
| y2kenny | I have been using the cleanup loop to periodically refresh the token. | 22:14 |
| y2kenny | anyway... I should probably think about this some more. | 22:14 |
| *** rfolco has quit IRC | 22:43 | |
| *** tosky has quit IRC | 22:49 | |
| *** rfolco has joined #zuul | 22:51 | |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: WIP: Scheduler: Reconfiguration ref-updated create/delete https://review.opendev.org/739198 | 23:07 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: WIP: Scheduler: Reconfiguration ref-updated oldrev+newrev https://review.opendev.org/739078 | 23:07 |
| *** hashar has quit IRC | 23:56 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!