| *** mattw4 has quit IRC | 00:00 | |
| *** jamesmcarthur has joined #zuul | 00:05 | |
| *** jamesmcarthur has quit IRC | 00:15 | |
| *** saneax has joined #zuul | 00:42 | |
| SpamapS | Having a weird problem trying to roll out a new secret | 00:46 |
|---|---|---|
| SpamapS | I have a playbook and secret defined in the same repo.. | 00:46 |
| SpamapS | and I'm changing the secret I'm pulling into a job from let's say "secret1" to "secret2" by changing it from {secrets: [secret1]} to {secrets: [{secret: secret2, name: secret1}]} .. | 00:47 |
| SpamapS | The problem I'm having now is that the content from secret1 is still being used in the gate job that changes this. | 00:47 |
| SpamapS | THe playbook is untrusted | 00:48 |
| SpamapS | so I thought it would use the new content int he gate | 00:48 |
| SpamapS | did I get the rules a little wrong? | 00:48 |
| *** igordc has quit IRC | 00:55 | |
| SpamapS | further info: the gate pipeline is marked post-review, but it does seem like it was using the old version of the config. | 01:19 |
| *** jamesmcarthur has joined #zuul | 01:24 | |
| SpamapS | (it's possible there was a PBKAC .. we force-pushed the commits and the promote pipeline jobs that use the same secret changes actually worked) | 01:30 |
| *** jamesmcarthur_ has joined #zuul | 01:32 | |
| *** jamesmcarthur has quit IRC | 01:34 | |
| *** saneax has quit IRC | 01:40 | |
| *** jamesmcarthur_ has quit IRC | 02:35 | |
| *** bhavikdbavishi has joined #zuul | 02:38 | |
| *** bhavikdbavishi1 has joined #zuul | 02:40 | |
| *** bhavikdbavishi has quit IRC | 02:42 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 02:42 | |
| *** michael-beaver has quit IRC | 02:54 | |
| openstackgerrit | Ian Wienand proposed zuul/nodepool master: [wip] dib_cmd https://review.opendev.org/672196 | 03:33 |
| openstackgerrit | Ian Wienand proposed zuul/nodepool master: [wip] dib_cmd https://review.opendev.org/672196 | 03:42 |
| *** bjackman_ has joined #zuul | 04:29 | |
| openstackgerrit | Ian Wienand proposed zuul/nodepool master: [wip] dib_cmd https://review.opendev.org/672196 | 04:34 |
| *** pcaruana has joined #zuul | 04:43 | |
| *** raukadah is now known as chandankumar | 04:44 | |
| *** AJaeger is now known as AJaeger_ | 05:02 | |
| *** threestrands has quit IRC | 05:04 | |
| *** saneax has joined #zuul | 05:15 | |
| *** bhavikdbavishi has quit IRC | 05:19 | |
| *** bhavikdbavishi has joined #zuul | 05:25 | |
| openstackgerrit | Ian Wienand proposed zuul/nodepool master: [wip] dib_cmd https://review.opendev.org/672196 | 05:35 |
| openstackgerrit | Ian Wienand proposed zuul/nodepool master: [wip] dib_cmd https://review.opendev.org/672196 | 05:46 |
| *** jamesmcarthur has joined #zuul | 05:55 | |
| *** jamesmcarthur has quit IRC | 05:56 | |
| *** igordc has joined #zuul | 06:08 | |
| *** jamesmcarthur has joined #zuul | 06:09 | |
| *** jhesketh has quit IRC | 06:11 | |
| *** jhesketh has joined #zuul | 06:11 | |
| *** jamesmcarthur has quit IRC | 06:13 | |
| *** igordc has quit IRC | 06:15 | |
| *** jamesmcarthur has joined #zuul | 06:16 | |
| *** jamesmcarthur_ has joined #zuul | 06:28 | |
| *** jamesmca_ has joined #zuul | 06:31 | |
| *** jamesmcarthur has quit IRC | 06:32 | |
| *** jamesmcarthur_ has quit IRC | 06:35 | |
| *** jamesmca_ has quit IRC | 06:38 | |
| *** jamesmcarthur has joined #zuul | 06:43 | |
| *** jamesmcarthur has quit IRC | 06:51 | |
| *** jamesmcarthur has joined #zuul | 06:53 | |
| *** bolg has joined #zuul | 07:03 | |
| *** bhavikdbavishi has quit IRC | 07:05 | |
| openstackgerrit | Ian Wienand proposed zuul/nodepool master: Add a dib_cmd option for diskimages https://review.opendev.org/672196 | 07:18 |
| *** zbr_ has joined #zuul | 07:20 | |
| *** irclogbot_1 has quit IRC | 07:20 | |
| *** openstackstatus has quit IRC | 07:20 | |
| *** irclogbot_2 has joined #zuul | 07:22 | |
| *** homeski has quit IRC | 07:23 | |
| *** zbr has quit IRC | 07:23 | |
| *** jamesmcarthur has quit IRC | 07:24 | |
| *** tosky has joined #zuul | 07:28 | |
| *** hashar has joined #zuul | 07:37 | |
| *** bhavikdbavishi has joined #zuul | 08:10 | |
| *** panda has quit IRC | 08:38 | |
| *** panda has joined #zuul | 08:38 | |
| *** jangutter has quit IRC | 09:01 | |
| *** jangutter has joined #zuul | 09:26 | |
| *** bhavikdbavishi has quit IRC | 09:59 | |
| *** shachar has joined #zuul | 10:29 | |
| *** snapiri has quit IRC | 10:32 | |
| mordred | corvus: I learned something while poking at your javascript change stack | 10:42 |
| mordred | corvus: when we raise 404 from cherrypy, it does not send cors headers (this is expected, we typically are setting them ont eh reponse after raising 404) | 10:43 |
| mordred | this means the javascript gets a CORS error rather than the actual error | 10:43 |
| mordred | from what I can tell, there is no mechanism in cherrypy to set headers while raising an httperror - although we CAN just set the status code in a normal response | 10:44 |
| openstackgerrit | Monty Taylor proposed zuul/zuul master: Don't barf in dashboard on CORS violations for 404s https://review.opendev.org/672262 | 10:50 |
| mordred | corvus: I *think* that should at least take care of the stacktrace | 10:50 |
| mordred | corvus: (you can see this by giving a local devserver dashboard an invalid build id) | 10:51 |
| *** sanjayu_ has joined #zuul | 10:58 | |
| *** saneax has quit IRC | 11:00 | |
| sshnaidm | why can I have error in zuul like "AttributeError: 'MergeJob' object has no attribute 'updated'" ? | 11:03 |
| sshnaidm | Zuul version: 3.9.1.dev64 | 11:03 |
| *** saneax has joined #zuul | 11:04 | |
| *** sanjayu_ has quit IRC | 11:06 | |
| *** hashar has quit IRC | 11:19 | |
| *** saneax has quit IRC | 11:22 | |
| webknjaz | @mordred: I can think of 2-3 ways of doing that in CherryPy. | 12:35 |
| mordred | webknjaz: woot! that's great news (I was hoping you'd say that :) ) | 12:35 |
| webknjaz | 1) `after_error_response` hook http://docs.cherrypy.org/en/latest/extend.html#hook-point | 12:36 |
| webknjaz | 2) `request.error_response` in config | 12:36 |
| webknjaz | 3) `error_page.default` in config | 12:36 |
| *** electrofelix has joined #zuul | 12:36 | |
| mordred | webknjaz: ooh yes. after_error_response seems like a great option - I like that a lot | 12:39 |
| Shrews | sshnaidm: do you have a traceback to share? | 12:41 |
| webknjaz | :+1 | 12:41 |
| sshnaidm | Shrews, well, after 2 restarts it's gone :/ | 12:41 |
| Shrews | sshnaidm: if you can capture that if you see it again, it would be helpful | 12:42 |
| mordred | webknjaz: it seems like if we're wanting to always set a given cors header, doing a tool that hooks on to before_finalize might also be a good choice, yes? | 12:43 |
| webknjaz | or on_end_resource | 12:44 |
| webknjaz | https://pypi.org/project/cherrypy-cors/ <-- from CherryPy maintainers | 12:45 |
| *** bjackman_ has quit IRC | 12:47 | |
| mordred | webknjaz: ooo. thanks | 12:49 |
| *** hashar has joined #zuul | 12:51 | |
| *** bhavikdbavishi has joined #zuul | 12:59 | |
| mordred | webknjaz: probably a dumb cherrypy question - but if I'm using the config method: https://github.com/yougov/cherrypy-cors/blob/master/cherrypy_cors.py#L41-L50 and I want to configure origins, do I just add a 'cors.expose.origins': 'foo' to the config? | 13:01 |
| webknjaz | yes | 13:02 |
| webknjaz | basically, all args of expose function you're referring to are directly mapped to the config | 13:02 |
| webknjaz | cors.expose.origins | 13:02 |
| *** bhavikdbavishi1 has joined #zuul | 13:02 | |
| webknjaz | cors.expose.allow_credentials | 13:02 |
| webknjaz | cors.expose.expose_headers | 13:02 |
| webknjaz | just like that | 13:03 |
| mordred | sweet | 13:03 |
| *** sshnaidm has quit IRC | 13:03 | |
| *** bhavikdbavishi has quit IRC | 13:04 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 13:04 | |
| openstackgerrit | Monty Taylor proposed zuul/zuul master: Use cherrypy_cors to set cors headers https://review.opendev.org/672285 | 13:05 |
| mordred | webknjaz: ^^ so potentially something like that | 13:05 |
| mordred | also - looking at the source of that, it's doing before_handler - and not after_error_response - my reading of the HTTPError exception is that it clears headers - does that mean I still need to do some plumbing to get it to run on after_error_response? (I'm thinking I could just add an after_error_response to the cors toolbox ...) | 13:08 |
| *** jamesmcarthur has joined #zuul | 13:08 | |
| webknjaz | I guess. I didn't write that lib | 13:09 |
| mordred | nod. mostly just thinking out loud - thanks for the pointers and feedback | 13:10 |
| webknjaz | I've also noticed `expose_public` tool there | 13:10 |
| mordred | yeah. just noticed that myself too - so I could do cors.expose_public.on instead of the two-line approach | 13:10 |
| webknjaz | https://github.com/yougov/cherrypy-cors/blob/master/cherrypy_cors.py#L188-L189 | 13:10 |
| webknjaz | just `expose` seems to be doing the request validation but not setting the origin | 13:11 |
| webknjaz | in only passes that arg to `_is_valid_origin` | 13:11 |
| webknjaz | so this only controls whether the incoming request hits your HTTP handler | 13:11 |
| *** sshnaidm has joined #zuul | 13:12 | |
| *** jamesmcarthur has quit IRC | 13:18 | |
| *** openstackstatus has joined #zuul | 13:22 | |
| *** ChanServ sets mode: +v openstackstatus | 13:22 | |
| openstackgerrit | Monty Taylor proposed zuul/zuul master: Use cherrypy_cors to set cors headers https://review.opendev.org/672285 | 13:24 |
| *** bhavikdbavishi has quit IRC | 13:35 | |
| *** jamesmcarthur has joined #zuul | 13:39 | |
| *** tosky_ has joined #zuul | 13:40 | |
| *** tosky has quit IRC | 13:42 | |
| *** jeliu_ has joined #zuul | 13:42 | |
| corvus | clarkb: can you +3 https://review.opendev.org/671874 when you get a chance? i'd like to get that into place so we can start seeing previews in the js build jobs | 13:42 |
| corvus | mordred: are the cors headers still going to be correct after the web auth stack finishes? | 13:43 |
| *** jeliu_ has quit IRC | 13:47 | |
| mordred | corvus: that's an excellent question | 13:48 |
| corvus | mordred: it looks like of the 3 that have landed so far, enqueue and dequeue had cors headers, and autohold did not. i don't know whether that's intentional, or accidental. | 13:48 |
| mordred | corvus: we can also use the cors tool as a decorator rather than as a global config | 13:48 |
| mordred | corvus: I think we should definitely do some more investigation before we land that - let me mark it WIP | 13:49 |
| *** AJaeger_ is now known as AJaeger | 13:50 | |
| openstackgerrit | Monty Taylor proposed zuul/zuul master: WIP Use cherrypy_cors to set cors headers https://review.opendev.org/672285 | 13:52 |
| *** jamesmcarthur has quit IRC | 13:58 | |
| *** michael-beaver has joined #zuul | 14:07 | |
| *** tosky_ is now known as tosky | 14:08 | |
| *** jeliu_ has joined #zuul | 14:18 | |
| sshnaidm | Shrews, you asked about traceback, just happened again: http://paste.openstack.org/show/754757/ | 14:21 |
| Shrews | interesting. that reference has been in the zuul code base since https://review.opendev.org/502989 | 14:27 |
| sshnaidm | Shrews, I found a few mentions of this error in internet, seems like it just happens because of misconfig or other issues | 14:51 |
| openstackgerrit | Monty Taylor proposed zuul/zuul master: WIP Use cherrypy_cors to set cors headers https://review.opendev.org/672285 | 15:01 |
| *** bhavikdbavishi has joined #zuul | 15:10 | |
| openstackgerrit | Monty Taylor proposed zuul/zuul master: WIP Do public cors without cherrypy_cors https://review.opendev.org/672313 | 15:12 |
| clarkb | corvus: I've approved the zuul generate manifests change | 15:30 |
| clarkb | corvus: I did leave a couple notes in there that were mostly observations if you want to take a look | 15:30 |
| clarkb | but nothing that said don't approve this as is | 15:30 |
| SpamapS | Is it possible that I'm getting a branch match problem? I have a job defined in master and prod, and when I update the job in master to use a different secret (but put it in the same variable), the job runs with the set of secrets from the prod branch version. | 15:45 |
| SpamapS | I'd have thought implied branch matchers would prevent this. | 15:45 |
| clarkb | (this is from fuzzy memory) I want to say secrets can't differ between branches? But maybe that was another data type | 15:47 |
| SpamapS | the actual secret doesn't | 15:48 |
| SpamapS | the reference to them is the change I'm making | 15:48 |
| SpamapS | In a gate job, I'm changing {secrets:[apikey]} to {secrets:[{secret: apikey2, name:apikey}]} | 15:49 |
| SpamapS | And in said gate job, it runs with the contents of apikey set to the old apikey secret, not the new one. | 15:49 |
| SpamapS | But other jobs in the promote pipeline run with the new content. | 15:50 |
| SpamapS | (those other jobs exist in both branches too) | 15:50 |
| SpamapS | It's kind of a bummer that secrets aren't mentioned in the UI. | 15:50 |
| SpamapS | definitely complicating debugging this | 15:51 |
| SpamapS | Oh also, I had a crazy idea that I will try to find time to submit as a patch. encrypt_secret.py needs to generate a sha256 of the plaintext and put that in to allow folks who have the secret, but not the encryption keys, to validate that the secret has the content they want. | 15:53 |
| SpamapS | (I've tested in our sandbox repo, and the secret updates properly even when the references on that job are different in prod.. so I'm kind of confused what's happening) | 16:00 |
| *** jeliu_ has quit IRC | 16:01 | |
| clarkb | SpamapS: has the change merged? | 16:02 |
| clarkb | Possible that subset of config is treated as trusted and must be post merge? | 16:03 |
| clarkb | this way you can't expose a different secret simply by changing the name? | 16:03 |
| SpamapS | Yes, the job is still using the wrong secret content even after I manually pushed the change in to master. | 16:03 |
| SpamapS | and restarted zuul-scheduler. | 16:03 |
| SpamapS | New changes still use the wrong secret. :-/ | 16:04 |
| SpamapS | and that's where I say the UI not mentioning secrets is causing me issues.. because it's hard to tell if Zuul knows about the change, or where it's getting the old one from. | 16:04 |
| *** bolg has quit IRC | 16:05 | |
| *** mattw4 has joined #zuul | 16:06 | |
| fungi | on the idea of publishing checksums of secret plaintext... that defeats one of the security guarantees (specifically that if two projects encrypt the same secret, they won't be able to tell) | 16:12 |
| SpamapS | fungi: ah that is a good reason not to do it. ;) | 16:12 |
| fungi | yes, i'll admit i had to think for several minutes about your suggestion to realize that | 16:13 |
| SpamapS | We were just struggling yesterday to debug. | 16:13 |
| SpamapS | And thought "would be nice if we had the sha or something" | 16:13 |
| SpamapS | Because now all we can do is autohold a node and actually look at the real secret. | 16:14 |
| fungi | i suppose if your secret contained some embedded comment/serial which updated on each iteration, you could intentionally leak that portion in logs... though that seems mildly risky and also could possibly enable certain kinds of known-plaintext attacks (though probably not) | 16:15 |
| SpamapS | http://paste.openstack.org/show/ByJ7ci23fV2zXmshKM7T/ | 16:16 |
| SpamapS | Here's the inheritance path | 16:16 |
| SpamapS | This makes very little sense to me. Why does both the master and prod variant match? | 16:16 |
| SpamapS | Like, I'd never looked at this inheritance thing.. but had you asked me before I saw this one, I'd say you should never see the same job with two variants in that chain. | 16:18 |
| clarkb | the branchmatcher is for master is that an explicit match? | 16:19 |
| clarkb | (that might actually be a good improvement to that debug tool, specifying if a match is implicit or explicit) | 16:20 |
| SpamapS | No there's no explicit branches statements anywhere in the config. | 16:20 |
| SpamapS | oh wait.. damnit | 16:21 |
| SpamapS | there is one explicit branches statement | 16:21 |
| SpamapS | on that stupid job | 16:21 |
| SpamapS | >:| | 16:21 |
| SpamapS | clarkb:well spotted | 16:21 |
| SpamapS | I went through and swept them all out a few months ago but I see how this one got added back in by some cargo-culting after that. | 16:23 |
| SpamapS | I almost wish I could disable explicit branch matching. | 16:24 |
| SpamapS | It's basically never what I want for our flow. | 16:24 |
| *** hashar has quit IRC | 16:32 | |
| openstackgerrit | Monty Taylor proposed zuul/zuul master: WIP Use cherrypy_cors to set cors headers https://review.opendev.org/672285 | 16:33 |
| openstackgerrit | Monty Taylor proposed zuul/zuul master: WIP Do public cors without cherrypy_cors https://review.opendev.org/672313 | 16:33 |
| corvus | 2019-07-23 15:34:42.811503 | ubuntu-bionic | ERROR: GitPython requires Python '>=3.0, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*' but the running Python is 2.7.15 | 16:36 |
| corvus | i guess that just changed | 16:36 |
| corvus | that's from the py27 jobs for zuul-jobs | 16:36 |
| fungi | gitpython 2.1.12 released two days ago | 16:38 |
| corvus | this is pretty weird. we need "zuul" installed for testing zuul-jobs, and of course it's py3 only. but zuul-jobs isn't -- we still want it to run under python2 since some of its ansible modules may run under py2 | 16:38 |
| corvus | i'm... not entirely sure how to resolve this | 16:39 |
| fungi | the changelog doesn't mention dropping 2.7 support | 16:39 |
| corvus | i guess one choice would be to go ahead and add a py2-only pin for gitpython in zuul, even though zuul itself isn't going to use it. another might be to avoid installing zuul via test-requirements, and instead install it standalone in some script called by tox.... | 16:40 |
| fungi | ahh, now their changelogs just link to version diff commit lists, and this is in it: https://github.com/gitpython-developers/GitPython/commit/dac619e4917b0ad43d836a534633d68a871aecca | 16:41 |
| fungi | "Drop python 2.7 support and help with encodings" | 16:41 |
| corvus | hrm, maybe if gitpython is unpinned in zuul, we can pre-install an old version in zuul-jobs and it will satisfy? | 16:42 |
| fungi | 2.1.11 should still work, yeah | 16:43 |
| fungi | https://github.com/gitpython-developers/GitPython/issues/898 points out the strangeness of dropping python 2 support in a patch revision | 16:43 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Add generate-zuul-manifest role https://review.opendev.org/671874 | 16:44 |
| corvus | fungi: indeed :) | 16:44 |
| corvus | fungi: ^ mabye that will work? | 16:44 |
| corvus | clarkb: replied, thanks | 16:47 |
| corvus | okay that looks like it worked | 16:53 |
| corvus | clarkb, mordred: can you re-review https://review.opendev.org/671874 please? one line change to pin gitpython | 16:53 |
| clarkb | corvus: you could use an env marker there to restrict the cap to python2 only if you want | 17:01 |
| corvus | clarkb: yeah, it doesn't really matter though, the code isn't actually run | 17:04 |
| corvus | (the only zuul code that might get run in those tests is the zuul_return module) | 17:05 |
| *** chandankumar is now known as raukadah | 17:08 | |
| clarkb | k I +2'd | 17:12 |
| openstackgerrit | Merged zuul/nodepool master: static: add host-key-checking toggle https://review.opendev.org/653679 | 17:23 |
| openstackgerrit | Merged zuul/zuul-jobs master: Add generate-zuul-manifest role https://review.opendev.org/671874 | 17:39 |
| *** panda is now known as panda|off | 17:45 | |
| *** sshnaidm is now known as sshnaidm|afk | 17:50 | |
| openstackgerrit | David Shrewsbury proposed zuul/nodepool master: Add build ID to failure message https://review.opendev.org/672337 | 17:52 |
| Shrews | easily pad those sweet sweet review stats ^^^ | 17:55 |
| Shrews | i might actually want to restart builders with that soon'ish | 17:55 |
| corvus | hey cool, it got dequeued from check :) | 18:14 |
| corvus | pipeline supercedence in action | 18:14 |
| corvus | and it failed tests :( | 18:14 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: DNM: Test base-jobs https://review.opendev.org/671894 | 18:15 |
| *** bhavikdbavishi has quit IRC | 18:19 | |
| *** bhavikdbavishi has joined #zuul | 18:19 | |
| Shrews | umm... how does one specify the tenant in the zuul status page? | 18:25 |
| clarkb | Shrews: Go to https://zuul.opendev.org | 18:25 |
| clarkb | it should give you a table to pick from | 18:25 |
| Shrews | ah, thx | 18:26 |
| *** hashar has joined #zuul | 18:26 | |
| *** igordc has joined #zuul | 18:27 | |
| Shrews | corvus: well those are fun, unrelated failures. neato | 18:27 |
| corvus | Shrews: yeh, looks like bad luck with zk | 18:41 |
| Shrews | "if it weren't for bad luck..." yadda yadda yadda | 18:47 |
| Shrews | corvus: before i recheck, how will that be handled in a supercedes pipeline? will it still go through check? | 18:47 |
| corvus | Shrews: yeah; i think we need to tweak that a little; maybe add "reverify" back | 18:48 |
| Shrews | or even "sudo gate" :) | 18:49 |
| corvus | haha | 18:49 |
| corvus | Shrews: a +W after it returns may put it back in gate | 18:49 |
| corvus | (i can't remember if we cleaned up the requirements for that or not) | 18:49 |
| Shrews | ok. i can add that myself | 18:50 |
| corvus | ++ worth a try | 18:50 |
| Shrews | ah neat. another +W worked | 18:51 |
| clarkb | ^^^ is why I -1'd that change to undo modern gerrit event behavior in zuul | 18:52 |
| corvus | i don't think Shrews had a +W on it before? | 18:52 |
| Shrews | i didn't | 18:53 |
| clarkb | ah that particular case would've worked then but if whoever had +W'd added a +W it would've worked too | 18:53 |
| *** bhavikdbavishi has quit IRC | 18:56 | |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Fix typo in generate-zuul-manifest role https://review.opendev.org/672343 | 18:58 |
| corvus | clarkb: ^ got a quick sec to +3 that? | 18:58 |
| corvus | (base-test scores again) | 18:58 |
| clarkb | done | 18:59 |
| *** hashar has quit IRC | 18:59 | |
| *** igordc has quit IRC | 19:00 | |
| openstackgerrit | Merged zuul/zuul-jobs master: Fix typo in generate-zuul-manifest role https://review.opendev.org/672343 | 19:11 |
| *** panda|off has quit IRC | 19:23 | |
| *** panda has joined #zuul | 19:25 | |
| openstackgerrit | Merged zuul/nodepool master: Add build ID to failure message https://review.opendev.org/672337 | 19:29 |
| *** tosky has quit IRC | 19:47 | |
| openstackgerrit | Monty Taylor proposed zuul/nodepool master: Install libffi6 on dpkg platforms https://review.opendev.org/672352 | 20:13 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Add log browsing to build page https://review.opendev.org/671906 | 20:27 |
| corvus | mordred: where did https://review.opendev.org/672352 crop up? | 20:40 |
| mordred | corvus: I just noticed it while using nodepool's dockerfile as an example to tell mnaser about python-builder/python-base | 20:40 |
| mordred | corvus: "why are we installing a dev library in a final install?" | 20:41 |
| corvus | gotcha | 20:41 |
| corvus | \o/ http://logs.openstack.org/06/671906/3/check/zuul-build-dashboard/a09ad90/npm/html/status will demonstrate the log work | 20:43 |
| clarkb | you have to drop the /status | 20:44 |
| corvus | sorry, it auto-redirects | 20:44 |
| corvus | http://logs.openstack.org/06/671906/3/check/zuul-build-dashboard/a09ad90/npm/html/ | 20:44 |
| corvus | open that up, go to the builds tab... then we need to find a build new enough to have a manifest | 20:45 |
| corvus | the airship jobs have it | 20:45 |
| corvus | you'll know if you see a "Logs" tab on the build page | 20:46 |
| clarkb | I'm not seeing airship builds? | 20:46 |
| corvus | scrolled off | 20:46 |
| clarkb | corvus: which project in airhsip I should be able to search that way | 20:46 |
| corvus | airship/airship-in-a-bottle | 20:47 |
| mordred | corvus: that's really nice | 20:48 |
| clarkb | corvus: quick feedback, it would be good to have a download original or similar type link too | 20:49 |
| corvus | clarkb: yes! that's actually in the spec | 20:49 |
| clarkb | great | 20:49 |
| corvus | the job-output.txt is working for me; some of the other text files aren't -- so make sure you try that one at least | 20:50 |
| clarkb | ya that one worked for me | 20:50 |
| *** pcaruana has quit IRC | 20:50 | |
| clarkb | I have to vim $url a lot of log files because browser chokes on them | 20:50 |
| clarkb | (and then there are the non text files) | 20:50 |
| corvus | yeah, this isn't going to win any prizes for efficiency; i think it might be nice to see if we can do some sort of progressive display here | 20:51 |
| corvus | but right now, it's totally 'read it all into memory, then add it all to the dom' | 20:51 |
| mordred | corvus: I'm sure there has to be a progressive display option :) | 20:53 |
| mordred | corvus: but I think starting with a thing that exists is awesome | 20:53 |
| corvus | i like how react-redux state works -- if you go back and forth between different files in the log viewer, it doesn't have to fetch the api info or manifest again | 20:55 |
| corvus | if you're eagle-eyed (or you have a slow connection) you may be able to see the logs tab appear after the builds page loads; that's because it doesn't show up until after the manifest is fetched and parsed | 20:56 |
| SpamapS | log viewer?! | 20:57 |
| * SpamapS has been mole-eyed and didn't spot that until now | 20:57 | |
| corvus | SpamapS: it's a wip, only exists in | 20:58 |
| corvus | https://review.opendev.org/671906 | 20:58 |
| SpamapS | sounds great tho | 20:58 |
| SpamapS | it's a common complaint of my users | 20:58 |
| SpamapS | "All this ANSI in the text logs confuses me." | 20:58 |
| mordred | SpamapS: I think it'll be a followup to translate all of the text in the logs to emoji | 21:00 |
| corvus | there must be a js library for that | 21:00 |
| mordred | corvus: are there any js libraries that AREN'T for that? | 21:00 |
| SpamapS | I assumed js was an emoji generation language. | 21:02 |
| SpamapS | https://imgur.com/AKFGrfP | 21:03 |
| mordred | wow. and with that - I'm going to go eat the foods | 21:04 |
| fungi | emojiaas | 21:08 |
| fungi | there is no ansi, only zuul | 21:09 |
| openstackgerrit | Merged zuul/nodepool master: Install libffi6 on dpkg platforms https://review.opendev.org/672352 | 21:15 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Add log browsing to build page https://review.opendev.org/671906 | 21:31 |
| corvus | i'm going to look into the docs build failure unless someone has started on that | 21:32 |
| *** irclogbot_2 has quit IRC | 21:32 | |
| *** altlogbot_0 has quit IRC | 21:33 | |
| fungi | i have not yet, sorry | 21:35 |
| *** altlogbot_2 has joined #zuul | 21:35 | |
| *** irclogbot_2 has joined #zuul | 21:35 | |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Fix sphinx error https://review.opendev.org/672372 | 21:44 |
| corvus | that should let us merge stuff | 21:44 |
| *** irclogbot_2 has quit IRC | 21:59 | |
| *** altlogbot_2 has quit IRC | 22:01 | |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Move artifacts to their own section https://review.opendev.org/672379 | 22:09 |
| *** altlogbot_0 has joined #zuul | 22:23 | |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Download-artifact: use the artifact type rather than name https://review.opendev.org/672381 | 22:26 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Use human-readable names for artifact returns https://review.opendev.org/672382 | 22:26 |
| *** altlogbot_0 has quit IRC | 22:27 | |
| *** altlogbot_2 has joined #zuul | 23:15 | |
| *** altlogbot_2 has quit IRC | 23:19 | |
| dmsimard | corvus: the manifest log browser is super dope | 23:19 |
| *** altlogbot_2 has joined #zuul | 23:29 | |
| *** irclogbot_2 has joined #zuul | 23:33 | |
| *** igordc has joined #zuul | 23:35 | |
| *** jamesmcarthur has joined #zuul | 23:36 | |
| *** jamesmcarthur has quit IRC | 23:52 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!