Wednesday, 2020-09-02

« Tuesday, 2020-09-01 Index Thursday, 2020-09-03 »
*** rlandy has quit IRC00:03
*** bolg has quit IRC00:04
*** flaper87 has quit IRC00:24
*** frenzyfriday has joined #zuul00:25
*** flaper87 has joined #zuul00:26
*** frenzyfriday has quit IRC00:33
*** frenzyfriday has joined #zuul02:14
*** frenzyfriday has quit IRC02:18
*** frenzyfriday has joined #zuul03:08
*** frenzyfriday has quit IRC03:12
*** bhavikdbavishi has joined #zuul03:42
*** bhavikdbavishi1 has joined #zuul03:47
*** bhavikdbavishi has quit IRC03:48
*** bhavikdbavishi1 is now known as bhavikdbavishi03:48
*** frenzyfriday has joined #zuul03:49
*** frenzyfriday has quit IRC03:59
*** bhavikdbavishi has quit IRC04:03
*** bhavikdbavishi has joined #zuul04:04
*** armstrongs has quit IRC04:05
*** evrardjp has quit IRC04:33
*** evrardjp has joined #zuul04:33
*** frenzyfriday has joined #zuul04:44
*** frenzyfriday has quit IRC04:54
*** wuchunyang has joined #zuul04:54
*** vishalmanchanda has joined #zuul04:55
*** bhagyashris|away is now known as bhagyashris05:27
*** frenzyfriday has joined #zuul05:38
*** frenzyfriday has quit IRC05:48
*** frenzyfriday has joined #zuul06:01
*** evgenyl has quit IRC06:03
*** evgenyl_ has joined #zuul06:03
*** mach1na has joined #zuul06:12
*** mach1na has quit IRC06:22
*** mach1na has joined #zuul06:23
tobiashclarkb: want to review https://review.opendev.org/747127? (adds zuul/zuul-client repo)06:29
*** bhavikdbavishi has quit IRC06:29
*** mach1na has quit IRC06:40
*** mach1na has joined #zuul06:49
*** armstrongs has joined #zuul07:00
*** armstrongs has quit IRC07:05
*** mach1na has quit IRC07:05
*** hashar has joined #zuul07:07
*** jcapitao has joined #zuul07:07
*** mach1na has joined #zuul07:08
*** bhavikdbavishi has joined #zuul07:19
*** bhavikdbavishi1 has joined #zuul07:26
*** bhavikdbavishi has quit IRC07:28
*** bhavikdbavishi1 is now known as bhavikdbavishi07:28
*** sshnaidm|afk is now known as sshnaidm07:32
*** tosky has joined #zuul07:37
*** jpena|off is now known as jpena07:55
*** holser has joined #zuul08:09
*** frenzyfriday has quit IRC09:12
mhuhello tobiash, looks like zuul-client wasn't created, is that normal? I can't clone it09:13
mhuoh, the deploy pipeline jobs just started, nevermind09:15
*** frenzyfriday has joined #zuul09:16
AJaegermhu: yes, you have to wait for the deploy job to run on the change - is all fine now?09:23
mhuAJaeger, yeah it's cloned nice and cozy on my system09:23
AJaegergreat - let's fill it with live ;)09:24
mhuyep, I'd like to get it started today09:24
AJaegerthanks, mhu !09:30
*** frenzyfriday has quit IRC09:34
*** bhavikdbavishi has quit IRC09:46
*** bhavikdbavishi has joined #zuul09:47
*** hashar has quit IRC09:53
*** nils has joined #zuul09:54
*** wuchunyang has quit IRC10:11
*** zenkuro has joined #zuul10:17
*** yolanda has joined #zuul10:26
*** bolg_ has joined #zuul10:30
*** bhavikdbavishi has quit IRC10:36
*** bhavikdbavishi has joined #zuul10:52
*** jcapitao is now known as jcapitao_lunch10:56
*** mach1na has quit IRC11:19
*** jpena is now known as jpena|lunch11:45
*** tflink_ has joined #zuul11:46
*** tflink has quit IRC11:46
*** Goneri has joined #zuul11:49
*** mach1na has joined #zuul11:53
*** mach1na has quit IRC11:57
*** mach1na has joined #zuul11:57
*** mattd01 has joined #zuul12:00
*** rlandy has joined #zuul12:07
*** vorotech has joined #zuul12:11
*** jcapitao_lunch is now known as jcapitao12:19
*** openstackgerrit has joined #zuul12:28
openstackgerritGuillaume Chauvel proposed zuul/zuul master: tutorial: Rework quick-start and prepare for other tutorials  https://review.opendev.org/73206612:28
openstackgerritGuillaume Chauvel proposed zuul/zuul master: tutorial: Add "gate your first patch"  https://review.opendev.org/73206712:28
openstackgerritGuillaume Chauvel proposed zuul/zuul master: tutorial: Add "Use zuul jobs"  https://review.opendev.org/73206812:28
openstackgerritGuillaume Chauvel proposed zuul/zuul master: tutorial: Add "gate pipeline"  https://review.opendev.org/73206912:28
openstackgerritGuillaume Chauvel proposed zuul/zuul master: tutorial: Add "job secrets"  https://review.opendev.org/73207012:28
openstackgerritGuillaume Chauvel proposed zuul/zuul master: tutorial: Add "job dependencies"  https://review.opendev.org/73207112:28
openstackgerritGuillaume Chauvel proposed zuul/zuul master: Rename quick-start to zuul-tutorial-quick-start  https://review.opendev.org/73765612:28
openstackgerritGuillaume Chauvel proposed zuul/zuul master: [DNM] TEST run zuul tutorials to test stream+callback (+ zuul-jobs change)  https://review.opendev.org/73547712:28
openstackgerritGuillaume Chauvel proposed zuul/zuul master: [DNM] Test: run multiple tutorials ('job dependencies' 2 times)  https://review.opendev.org/74155812:28
*** jpena|lunch is now known as jpena12:48
zbrianw: https://review.opendev.org/#/c/748737/ us fixing tox module on two platforms: centos-7 and debian-stretch12:56
*** frenzyfriday has joined #zuul13:07
*** johanssone has quit IRC13:11
*** fdegir has quit IRC13:11
tobiashclarkb: I addressed your comments in https://review.opendev.org/747142 (annotate getBranch)13:26
tristanCclarkb: no sorry, i don't know why openshift wheels are not published13:32
*** bhavikdbavishi has quit IRC13:32
*** sshnaidm is now known as sshnaidm|bbl13:58
*** vorotech has quit IRC14:09
fungialso that question could have been meant as a subtle hint to convince people to start publishing some14:11
*** zenkuro has quit IRC14:30
*** fdegir has joined #zuul14:34
*** johanssone has joined #zuul14:36
AJaegerzuul-jobs maintainers, please review this consolidation stack of upload log roles: https://review.opendev.org/#/c/74273214:51
*** frenzyfriday has quit IRC14:53
*** frenzyfriday has joined #zuul14:59
*** frenzyfriday has quit IRC15:03
zbrfungi: tristanC: please help me make cmd visible by default with https://review.opendev.org/#/c/740733/15:07
zbrusers should not have to press the magic button to see what was the command that failed, https://sbarnea.com/ss/Screen-Shot-2020-09-02-16-07-38.72.png15:08
*** frenzyfriday has joined #zuul15:31
*** tosky has quit IRC15:31
*** mach1na has quit IRC15:33
*** frenzyfriday has quit IRC15:36
*** hashar has joined #zuul15:48
*** mach1na has joined #zuul15:48
*** mach1na has quit IRC15:53
clarkbtobiash: on https://review.opendev.org/#/c/747142/1..2/zuul/driver/github/githubconnection.py the first comment I'm not sure that is correct still. I was pointing out that the event there is an actual event object and not an event id string. So we need to do =event.zuul_event_id or whatever the field is16:07
clarkbtobiash: see line 416 in that diff16:08
tobiashclarkb: technically it doesn't matter since the log adapter is explicitly designed to work with either an event object or directly the string16:09
tobiashclarkb: see https://opendev.org/zuul/zuul/src/branch/master/zuul/lib/logutil.py#L2616:11
tobiashit can be either the event id directly or something that contains zuul_event_id16:11
clarkbgot it16:12
tobiashbut I can change it for consistency if you want16:12
clarkbthere was another chang ethat normalized on the event_id I think which is probably why I had in my head that this is the "proper" method16:12
*** bhavikdbavishi has joined #zuul16:17
*** jcapitao has quit IRC16:31
*** nils has quit IRC16:41
openstackgerritMatthieu Huin proposed zuul/zuul master: REST API: improve tenant scoping of autohold, authorizations  https://review.opendev.org/72811816:44
openstackgerritMatthieu Huin proposed zuul/zuul master: CLI: add autohold-info, autohold-delete via REST  https://review.opendev.org/72841016:44
openstackgerritMatthieu Huin proposed zuul/zuul master: REST API: improve promote endpoint testing  https://review.opendev.org/72848916:44
openstackgerritMatthieu Huin proposed zuul/zuul master: Builds: add "held" attribute  https://review.opendev.org/72899416:44
*** jpena is now known as jpena|off17:00
openstackgerritMerged zuul/zuul-jobs master: Remove dependency on pkg_resources  https://review.opendev.org/74873717:06
openstackgerritMerged zuul/zuul-website master: Add new Supporter Logos  https://review.opendev.org/74298417:08
openstackgerritMerged zuul/zuul-website-media master: Add 3D logo  https://review.opendev.org/68694917:12
*** mgoddard has quit IRC17:12
mhuoooh fancy17:13
*** mattd01 has quit IRC17:16
fungithat's all tristanC's work, i was revisiting web site patches and realized it was still sitting there waiting for more reviews17:18
fungiand yes, very pretty17:18
*** vorotech has joined #zuul18:08
*** mattd01 has joined #zuul18:11
openstackgerritMerged zuul/zuul master: Improve rendering of cmd  https://review.opendev.org/74073318:23
openstackgerritMerged zuul/zuul master: Fix typo in documentation  https://review.opendev.org/74871618:23
openstackgerritMerged zuul/zuul master: Fix typos in inline comments  https://review.opendev.org/74907018:23
*** bhavikdbavishi has quit IRC18:25
*** hashar has quit IRC18:34
*** y2kenny has joined #zuul18:42
y2kennyWith the SMTP reporter, is it possible to send email to the committer or author's email address of the change?18:43
*** vishalmanchanda has quit IRC18:43
fungiy2kenny: i think zuul would need a new feature to be able to plumb that through to a pipeline variable... also i'm not entirely sure the scheduler (which is what handles reporting decisions) knows those commit header values18:58
fungii guess it's probably incorporated into event metadata we get for triggers, so could probably generalize the fields if they're present18:59
y2kennyfungi: ok thanks for the info.  I noticed there are some undocumented variables like {change.patchset} that can be use in the pipeline yaml.  I was wondering if the commit/author email address may be extracted.18:59
fungiy2kenny: yeah, i expect we'd need to add the addresses (for gerrit there are additional addresses too... change owner, reviewers...)19:00
funginot that i'm saying if one driver got the feature we'd need it merged at the same time for all of them, but knowing what sorts of addresses different code review platforms provide could help inform design decisions19:01
y2kennyfungi: so is "event metadata" the proper term for all those {variables} available in the yaml?19:02
y2kennyI only discovered them in some of the test cases in zuul19:02
y2kenny(for creating status url, etc.)19:02
fungigreat questions, i don't actually know for sure. zuul was designed around gerrit first, and gerrit's event stream contains "events" with "metadata"19:02
fungiso i sort of expect we kept to that terminology, but would need to go digging in the source to be sure19:03
fungialso i agree the lack of documentation around the reporter variables is a problem. i want to say the last time i brought it up there was a suggestion that they weren't documented because we wanted to deprecate them, but we've seen a lot of new use cases crop up and some of them might really expect to make use of that so it's worth revisiting either way19:04
y2kennylooks like change.number and change.patchset is available for at least pagure and gitlab19:05
y2kennyhttps://opendev.org/zuul/zuul/src/branch/master/tests/fixtures/zuul-gitlab-driver.conf#L519:05
y2kennyhttps://opendev.org/zuul/zuul/src/branch/master/tests/fixtures/zuul-pagure-driver.conf#L519:05
y2kennythis status_url is definitely useful19:05
fungijust by way of clarification, originally zuul did not have a builds dashboard and keep its own result history, so we relied on it posting review comments with links to job logs. to do that, we needed to be able to recreate the urls where those logs were expected to have been saved. all of that we hoped was obsoleted by adding a dashboard with retained state19:06
fungibut i will say, you're not the first person to ask whether they could have zuul send build result reports via smtp reporter to (author, committer, owner, approver...) addresses19:08
y2kennyso in my case, I also used it to customize the start-message... although I haven't been seeing start-message being reported so may be my usage my usage pattern broke after update19:09
y2kenny"Starting {pipeline.name} jobs.  Status:\n{status_url}/change/{change.number},{change.patchset}"19:09
*** y2kenny has quit IRC19:10
fungifor context, the origins of the smtp reporter were that we wanted to be able to report build results/failures for periodic pipeline builds to specific mailing lists, and then people who were interested in those could subscribe to the mailing lists (outside zuul) to receive those reports. this was again because there was no builds history/dashboard and since they weren't triggered by changes there was no change19:11
fungifor zuul to leave comments on19:11
fungiwe hoped by providing a queryable dashboard with job build histories, the smtp reporter would be less necessary19:13
*** vorotech has quit IRC19:13
*** y2kenny has joined #zuul19:13
fungiy2kenny: http://eavesdrop.openstack.org/irclogs/%23zuul/%23zuul.2020-09-02.log.html#t2020-09-02T19:11:2519:17
fungifor what i was saying while you were offline19:17
y2kennyfungi: thanks.  I was looking at it but the eavesdrop log hasn't updated.19:18
fungithe html gets rebuilt via a cron job, you can go to http://eavesdrop.openstack.org/irclogs/%23zuul/%23zuul.2020-09-02.log for the raw version19:19
fungi(just drop the .html extension off the url)19:20
fungithat should update roughly continuously, or at least as often as the bot flushes its output to disk19:20
y2kennyfungi: so for the smtp usecase I am looking at is for people who don't want to use anything to review except for email.  The use case is kind of problematic in general anyway (like... one can push a commit that they are neither the author or committer.)19:20
y2kennyah ok... that's good to know about the log19:20
fungiy2kenny: yep, i think that use case makes sense, we just need to design for it19:21
fungiand so far we haven't, as far as i know19:21
fungifor those people so far we've mostly relied on the fact that many (most?) code review platforms send e-mail comments, or can be configured to do so19:22
y2kennyyup and that's what I will probably say to them for now... but there are folks who want to push to refs/heads/* instead of refs/for/*19:22
fungifor example gerrit sends me e-mail for every comment left on a change i started (where i'm the change "owner" in its parlance) so any result reports zuul posts to those changes get e-mailed to me by gerrit19:23
fungiand yeah, i get you, they're not actually using code review then ;)19:23
*** tflink_ is now known as tflink19:23
y2kennyyea... I wish everyone just use Gerrit :)19:23
fungimordred and corvus basically insisted that all openstack commits would go through code review, which was quite likely the best decision in the history of that project, and is what really allowed us to build workflows which eventually evolved to become zuul19:25
*** Eighth_Doctor has joined #zuul19:46
*** frenzyfriday has joined #zuul19:54
*** frenzyfriday has quit IRC20:02
*** frenzyfriday has joined #zuul20:14
*** y2kenny has quit IRC20:23
*** frenzyfriday has quit IRC20:24
*** vorotech has joined #zuul20:44
*** frenzyfriday has joined #zuul20:49
*** dustinc has joined #zuul20:58
*** frenzyfriday has quit IRC21:05
clarkbtobiash: https://review.opendev.org/#/c/651526/9 is a neat feature. I'm sorry it too me so long to ever review it.21:17
tobiashyeah, we want to use it for notifying users about problems with the job like full disk :)21:22
*** Goneri has quit IRC21:30
*** vorotech has quit IRC21:31
*** piotrowskim has quit IRC21:33
clarkbmhu: left a note on https://review.opendev.org/#/c/728994/18 the db (and sqlalchemy) stuff is something I won't claim expertise on so I dind't -1 but if my concern is valid that may be worthy of a -121:52
*** SpamapS has quit IRC21:59
*** paladox has quit IRC21:59
*** arxcruz|ruck has quit IRC21:59
*** EmilienM has quit IRC21:59
*** paladox has joined #zuul22:00
*** arxcruz has joined #zuul22:00
clarkbtobiash: tomorrow do you think you can review https://review.opendev.org/#/c/738013/ ? then if that lands I can rebase my change that is based on it22:00
*** EmilienM has joined #zuul22:00
*** SpamapS has joined #zuul22:00
clarkb(only my change is merge conflicted so I don't want to rebase both)22:00
*** mattd01 has quit IRC22:26
*** rlandy has quit IRC22:34
*** sshnaidm|bbl is now known as sshnaidm|afk22:47
clarkbtristanC: I +2'd https://review.opendev.org/#/c/716298/3 but didn't approve it as I don't have a good way to test that and we don't use those example base jobs in opendev anyway. Feel free to approve when you are able to confirm its happy22:52
*** dustinc has quit IRC23:08
*** armstrongs has joined #zuul23:29
*** irclogbot_1 has quit IRC23:29
*** irclogbot_1 has joined #zuul23:34
« Tuesday, 2020-09-01 Index Thursday, 2020-09-03 »

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!