| *** hasharAway has quit IRC | 00:23 | |
| *** jasondotstar has joined #zuul | 00:27 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: zk: use client.retry instead of direct call https://review.openstack.org/523640 | 01:29 |
|---|---|---|
| tristanC | jeblair: would be nice to have zuul-jobs releases, at least to snapshot a known set of roles and jobs | 01:38 |
| *** jasondotstar has quit IRC | 01:41 | |
| *** jasondotstar has joined #zuul | 01:45 | |
| tristanC | pabelanger: re zuul-dashboard, moving the html/js bits to a dedicated repos has been "discussed" here http://lists.openstack.org/pipermail/openstack-infra/2017-July/005515.html | 02:20 |
| tristanC | well i suggested to keep the files in-repos because they are served by zuul-web, and i guess we'll wait for better javascript tooling to move the code out and make it easily consumable by zuul-web | 02:21 |
| tristanC | ideally, we would have a python-zuulclient and a zuul-webclient, similarly to storyboard structure | 02:24 |
| pabelanger | tristanC: yah, I like the idea of eventually moving it into own repo (maybe even now). I like the idea of python / javascript split on repos | 02:25 |
| tristanC | pabelanger: i like the idea too, the zuul-web static code would be easy to refactor, but i'm not sure how to create a standalone javascript project | 02:28 |
| pabelanger | tristanC: I suspect mordred might have some thoughts :) | 02:28 |
| *** threestrands has joined #zuul | 03:05 | |
| *** jaianshu has joined #zuul | 03:48 | |
| *** dmellado has quit IRC | 04:13 | |
| *** dmellado has joined #zuul | 04:15 | |
| *** threestrands has quit IRC | 04:29 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: zk: use client.retry instead of direct call https://review.openstack.org/523640 | 04:29 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Refactor provider config to driver module https://review.openstack.org/488384 | 04:37 |
| *** threestrands has joined #zuul | 04:41 | |
| *** threestrands has quit IRC | 04:41 | |
| *** threestrands has joined #zuul | 04:41 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: github: add integration documentation https://review.openstack.org/522420 | 05:17 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Refactor provider config to driver module https://review.openstack.org/488384 | 05:22 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: mqtt: add basic reporter https://review.openstack.org/518279 | 05:28 |
| *** threestrands has quit IRC | 05:32 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: web: add Cache-Control to static files https://review.openstack.org/522163 | 05:32 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: web: add Cache-Control to static files https://review.openstack.org/522163 | 06:11 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: web: add /{tenant}/jobs route https://review.openstack.org/503270 | 06:11 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: web: add /{tenant}/builds route https://review.openstack.org/466561 | 06:11 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: web: make console-stream tenant scoped https://review.openstack.org/505452 | 06:11 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: web: add /{source}/{project}.pub route https://review.openstack.org/502530 | 06:11 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: web: add Cache-Control to static files https://review.openstack.org/522163 | 06:20 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: web: add Cache-Control to static files https://review.openstack.org/522163 | 06:22 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: executor: add log_stream_port and log_stream_file settings https://review.openstack.org/523697 | 06:23 |
| tristanC | tobiash: weren't you looking for ways to make zuul_console connect back to the executor zuul_stream module? | 06:25 |
| tristanC | tobiash: well i tried first to be able to at least change the port of the zuul_console server and couldn't find a better way than in 523697 | 06:25 |
| tristanC | tobiash: and i was thinking adding a connect_back boolean that could be set the same way to enable the zuul_console -> zuul_stream flow | 06:26 |
| tristanC | i'm interested in a connect back because it's going to be hard to route fixed port inbound traffic to a hypothetical k8s slave | 06:29 |
| *** tushar has quit IRC | 06:35 | |
| *** tushar has joined #zuul | 06:35 | |
| tobiash | tristanC: I didn't look into this topic | 06:36 |
| tobiash | tristanC: but for the port change and possibly connect back | 06:37 |
| tobiash | tristanC: the zuul_console is an ordinary ansible module | 06:37 |
| tobiash | tristanC: so I think you might be able to add optional arguments to this to configure that instead of patching the module | 06:37 |
| tobiash | tristanC: this way your base job could configure it | 06:38 |
| tristanC | tobiash: the issue is to make the zuul_console module port match the zuul_stream callback, hence that idea to patch both in the same place | 06:39 |
| tobiash | tristanC: ah, missed that point | 06:39 |
| tobiash | hrm | 06:39 |
| tristanC | tobiash: my bad, for some reason i though you already looked into connect back | 06:39 |
| tobiash | I also think there was a discussion about that, maybe jlk or SpamapS discussed that | 06:41 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: github: add integration documentation https://review.openstack.org/522420 | 06:43 |
| tobiash | tristanC: how do you connect to your hypothetical k8s slave? | 06:43 |
| tobiash | you also need to route port 22 to it | 06:44 |
| tobiash | or do you use random ports for that? | 06:44 |
| tristanC | tobiash: the driver i proposed (whcih should better be renamed k8s-ssh) is using a random ssh port that is exposed on the kubelet with hostPort | 06:46 |
| tristanC | but i guess any other non ssh driver would have the same issue with zuul_stream trying to connect to zuul_console | 06:47 |
| tobiash | tristanC: your zuul_console port change makes the port configurable globally, wouldn't you need a per job setting? | 06:48 |
| tristanC | tobiash: yes, that's just a setting operator might want to change | 06:48 |
| tobiash | ah, ok | 06:48 |
| tobiash | tristanC: not sure how easy it is but this is ansible code so how about making this configurable via ansible.cfg instead of patching? | 06:49 |
| tobiash | that way you could make this setting globally and even easily configurable per job as ansible.cfg is generated per job afaik | 06:49 |
| tristanC | tobiash: oh, that may work indeed... i don't know enough ansible to say for sure though | 06:50 |
| tobiash | me 2 | 06:50 |
| tristanC | tobiash: what i'm trying to do right now is to spawn a single zuul_console on a oci "hypervisor", and make the log_stream_file shared between all the containers and the zuul_console daemon | 06:50 |
| tobiash | tristanC: so you have one zuul_console serving every console log | 06:54 |
| tobiash | ah no, you have one per k8s node as a DeamonSet? | 06:55 |
| tristanC | yes, one zuul_console per host, that would fix issue i'm having when concurrent container run in parralel, when the initial zuul_console get killed, the other job are getting "Waiting on logger" | 06:56 |
| tobiash | ah and you connect to the console log and ssh to the host ip and not the pod ip | 06:57 |
| tobiash | and that's why you need to change also the stream file | 06:59 |
| tobiash | now I get it | 06:59 |
| tobiash | but that global setting requires that you have a k8s-only zuul right? | 07:00 |
| tristanC | i would use the podIp if possible, but i couldn't figure out how to route the docker0 (172.17.0.1/16) network to the zuul-executor host, it seems like it's behind KUBE-FIREWALL | 07:01 |
| tristanC | and to support public k8s provider, I guess this won't work either, so that's why i think k8s would need zuul_console to connect back to zuul_stream | 07:02 |
| tristanC | for oci, it's another story, the driver doesn't set network namespace, so it's more simple, and when the driver start it could simply spawn a long live zuul_console daemon on the host | 07:03 |
| *** sshnaidm|off is now known as sshnaidm | 07:57 | |
| *** sshnaidm is now known as sshnaidm|afk | 08:11 | |
| *** hashar has joined #zuul | 08:15 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Refactor provider config to driver module https://review.openstack.org/488384 | 08:18 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Implement a static driver for Nodepool https://review.openstack.org/468624 | 08:18 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Implement an OpenContainer driver https://review.openstack.org/468753 | 08:18 |
| tristanC | fwiw, last PS of 468753 does spawn a single zuul_console per container host using a shared /var/tmp/zuul-console directory | 08:30 |
| *** bhavik1 has joined #zuul | 08:44 | |
| *** bhavik1 has quit IRC | 08:48 | |
| *** jaianshu has quit IRC | 10:03 | |
| *** electrofelix has joined #zuul | 10:22 | |
| *** sshnaidm|afk is now known as sshnaidm | 10:47 | |
| *** hashar is now known as hasharAway | 10:50 | |
| *** haint_ has joined #zuul | 11:16 | |
| *** haint has quit IRC | 11:21 | |
| *** jkilpatr has quit IRC | 11:46 | |
| *** jkilpatr has joined #zuul | 12:17 | |
| *** neatherweb has joined #zuul | 12:21 | |
| *** sshnaidm is now known as sshnaidm|afk | 12:22 | |
| neatherweb | hi - I'm trying to setup Zuul to see if I can use it for my CI, but currently getting errors with bwrap "Creating new namespace failed". Is there a version of bubblewrap I should use ? (I'm using Centos7) | 12:28 |
| tobiash | neatherweb: on centos you have to obey some things | 12:29 |
| tobiash | I assime centos 7.4? | 12:29 |
| neatherweb | yes | 12:29 |
| tobiash | neatherweb: you have to enable unprivileged namespaces: | 12:30 |
| tobiash | grubby --args="namespace.unpriv_enable=1" --update-kernel="$(sudo grubby --default-kernel)" | 12:30 |
| tobiash | and put the following into /etc/sysctl.d/98-user-namespaces.conf : | 12:30 |
| tobiash | user.max_user_namespaces=15076 | 12:30 |
| tobiash | then it should work | 12:30 |
| neatherweb | tobiash: great, thx. will give it a go | 12:31 |
| tobiash | I also struggled with this ;) | 12:31 |
| tristanC | tobiash: actually you don't have to enable usernamespace, it's disabled by default because this may be not production ready... | 12:41 |
| tobiash | tristanC: how can I then use the bwrap driver? | 12:42 |
| *** threestrands has joined #zuul | 12:42 | |
| *** threestrands has quit IRC | 12:42 | |
| *** threestrands has joined #zuul | 12:42 | |
| tristanC | tobiash: using this https://github.com/projectatomic/bubblewrap/commit/ec5093d57d8d55aa49525e26117ff4e43181a4d3 :-) | 12:42 |
| tristanC | which is part of the 0.2.0 release | 12:42 |
| tobiash | lemme check which version I have | 12:43 |
| tobiash | at least bwrap failed for me two days ago in my centos based openshift | 12:43 |
| tobiash | alpine 3.6 is still on 0.1.8 :( | 12:45 |
| tristanC | ah, then enabling userns works too :) | 12:46 |
| tobiash | well enabling userns is some extra manual deployment step in my openshift cluster | 12:46 |
| tobiash | if I could avoid that that would be great | 12:47 |
| tristanC | tobiash: couldn't you recompile bwrap with that patch? | 12:47 |
| tobiash | probably yes | 12:48 |
| tristanC | it's a silly behavior change, there used to be a /sys/module/user_namespace/parameters/enable file bwrap would read to tell if it could use userns. recent kernel removed that file and instead use user.max_user_namespaces sysctl to indicate userns are disabled | 12:50 |
| neatherweb | tobiash: thx, my 1st build job completes now :) | 12:56 |
| tobiash | tristanC: I think I can install it from the edge repo | 12:57 |
| tobiash | :) | 12:57 |
| tristanC | tobiash: or feel free to rpm2cpio https://softwarefactory-project.io/kojifiles/repos/sf-2.7-el7-release/Mash/bubblewrap-0.1.8-2.el7.x86_64.rpm | 12:58 |
| tobiash | tristanC: thanks, I'm using zuul in an alpine container on centos ;) | 12:58 |
| tobiash | but I've just found out how to install bwrap from the unrelease next alpine which is version 2.0 | 12:59 |
| tristanC | or you could probably trick bwrap 0.1.8 by mounting a tmpfs on /sys/module/user_namespace/parameters/ and echo 0 > enable | 12:59 |
| tristanC | anyway, you can use bwrap on recent el7 kernel without enabling userns | 13:00 |
| *** threestrands has quit IRC | 13:01 | |
| *** sshnaidm|afk is now known as sshnaidm | 13:26 | |
| *** openstack has joined #zuul | 13:46 | |
| *** ChanServ sets mode: +o openstack | 13:46 | |
| *** dkranz has joined #zuul | 14:21 | |
| mordred | tristanC: heya - sorry, I'm behind on the javascript stuff | 14:34 |
| mordred | tristanC, tobiash: oh - also, I have a todo list item that i probably won't get to until next week - but I've got a theoretical plan which should make it possible to get rid of zuul_console | 14:35 |
| *** openstack has quit IRC | 14:46 | |
| *** openstack has joined #zuul | 14:52 | |
| *** ChanServ sets mode: +o openstack | 14:52 | |
| *** sshnaidm is now known as sshnaidm|rover | 15:16 | |
| pabelanger | tobiash: tristanC: Wow, we should maybe write that down someplace. I haven't needed to change anything on fedora, which is nice. | 15:34 |
| pabelanger | mordred: do you have some time this morning to review: https://review.openstack.org/521324/ that is the rework of inventory file for a shared ansible_host setting | 15:49 |
| mordred | pabelanger: it is open | 16:07 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Fix branch checkout order https://review.openstack.org/523929 | 16:46 |
| openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: Combine branch templates and pipeline branch matchers https://review.openstack.org/523544 | 16:54 |
| *** weshay is now known as weshay_mtg | 17:00 | |
| *** sshnaidm|rover is now known as sshnaidm|rover|m | 17:10 | |
| *** sshnaidm|rover|m is now known as sshnaidm|rover | 17:10 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Normalize daemon process handling https://review.openstack.org/517381 | 17:10 |
| jeblair | mordred: can you decide if https://review.openstack.org/499224 is still necessary? | 17:28 |
| pabelanger | jeblair: mordred: are we using a specific topic for zuulv3.0 release in gerrit? | 17:35 |
| jeblair | pabelanger: no | 17:35 |
| pabelanger | k, in that case, the stack at https://review.openstack.org/523197 is ready for reviews. Adds commandsocket support | 17:35 |
| mordred | jeblair: based on jlk's comment I think no? | 17:38 |
| openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: Use username from node information if available https://review.openstack.org/453983 | 17:38 |
| mordred | jlk: ^^ around? thoughts? | 17:38 |
| jlk | Sorry I'm headed to dentist. I can look later? | 17:39 |
| openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: web: add Cache-Control to static files https://review.openstack.org/522163 | 17:41 |
| * SpamapS will peek | 17:44 | |
| jeblair | tobiash: question on 513331 | 17:50 |
| *** qwc has quit IRC | 17:54 | |
| *** qwc has joined #zuul | 17:56 | |
| openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: Re-enable test_zuul_trigger_project_change_merged https://review.openstack.org/513267 | 17:57 |
| openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: Remove unused function toList from scheduler https://review.openstack.org/513268 | 17:57 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Normalize daemon process handling https://review.openstack.org/517381 | 18:03 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Honor constraints files in tox-siblings https://review.openstack.org/523949 | 18:06 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Combine tox-siblings and tox roles https://review.openstack.org/523950 | 18:06 |
| jeblair | ianw: question on 519222 | 18:11 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/zuul-jobs master: build-javascript-tarball: add files to run job when needed https://review.openstack.org/523960 | 18:37 |
| clarkb | reading zuul docs required-projects is a job attribute, does that mean I can apply it at the project-template level and have it apply to all jobs in the template? | 18:39 |
| pabelanger | EmilienM: left a comment on 523960 | 18:41 |
| clarkb | grepping around looks like you have to set it at the job level in a template | 18:41 |
| EmilienM | pabelanger: ok, gotcha | 18:41 |
| EmilienM | pabelanger: so I need to do that in project-config, right? | 18:41 |
| pabelanger | EmilienM: no, you could carry a project-template your project zuul.yaml | 18:42 |
| EmilienM | pabelanger: ok | 18:42 |
| *** weshay_mtg is now known as weshay | 18:42 | |
| EmilienM | pabelanger: but I thought we shouldn't have standard jobs in projects | 18:43 |
| EmilienM | and rather do it in project-config | 18:43 |
| pabelanger | EmilienM: lets move to #openstack-infra, you are right, it might be a policy issue | 18:43 |
| pabelanger | not technical | 18:43 |
| jeblair | clarkb: your revised understanding is correct | 18:54 |
| clarkb | jeblair: https://review.openstack.org/523963 is why I asked | 18:55 |
| tobiash | jeblair: have to check again | 18:57 |
| jeblair | clarkb: 2 alternatives can make that better: 1) make legacy-puppet-beaker-rspec-infra inherit from legacy-puppet-beaker-rspec and add those projects. 2) use yaml anchors to avoid repeating the block. | 18:58 |
| jeblair | clarkb: or, of course, we could just go with what you have. at least it's in a template. :) | 18:58 |
| clarkb | re the first thing, the template is used all over the place so I'd have to update those to do 1) I think. 2) is actually not a half bad idea but this is also legacy so I'm not feeling like its necessary >_> | 18:59 |
| jeblair | clarkb: re 1 -- well, i was thinking you'd just update the template to use that new job. so it's 3 lines in the template that need changing after making the job. but now that i think about this further -- is that template infra-specific or generic? | 19:00 |
| jeblair | clarkb: if it's generic, then we'll be adding those projects to a bunch of puppet modules, yeah? | 19:01 |
| jeblair | clarkb: (with your change as written) | 19:01 |
| clarkb | jeblair: it appears to be infra specific because not infra uses the centos only job | 19:01 |
| jeblair | k | 19:01 |
| clarkb | so ya I guess I could've just added a new job that inherited and set that or just updated the existing job as is | 19:01 |
| jeblair | clarkb: ok. +2; i'll leave it to you to +W or revise. | 19:02 |
| clarkb | I think I'll keep it as is for now that way if say puppet openstck wanted to test on xenial again they could use the job without getting all of our modules because we've confined the madness to that specific template | 19:03 |
| clarkb | oh except if we have to make chagnes we may miss the three locations so let me collapse it into one location | 19:04 |
| *** electrofelix has quit IRC | 19:04 | |
| clarkb | jeblair: and does every job need to specify its own run even if it just wants to run its parents run playbook and call it good? | 19:08 |
| jeblair | clarkb: nope, run is a typical inherit/override attribute. so omit it and you get parent's run | 19:09 |
| clarkb | ok new patchset is up that is hopefully more maintainable | 19:10 |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul feature/zuulv3: Fix mixed canonical/non-canonical project merge https://review.openstack.org/513331 | 19:34 |
| tobiash | jeblair: you're right, updated ^^ | 19:35 |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul feature/zuulv3: Add regex support to project stanzas https://review.openstack.org/513368 | 19:36 |
| *** hasharAway is now known as hashar | 19:56 | |
| *** sshnaidm|rover is now known as sshnaidm|off | 20:01 | |
| clarkb | PTG planning is happening and they are asking me if the infra team (and by extension zuul) will be attending the PTG in dublin. I've only got until the 5th to respond so figure I will start an informal poll here | 20:02 |
| clarkb | please let me know if you plan to attend (or not) so that I can give them accurate info | 20:03 |
| dmsimard | I'm not going to Dublin :( I'll try to go to Vancouver though. | 20:03 |
| clarkb | (I understand that travel budgets and stuff won't eb sorted out yet so give me your best guess :) ) | 20:04 |
| Shrews | I'd like to, but my travel budget seems to be much more limited than others within the same company. I'll have to inquire about it. | 20:04 |
| tobiash | I'll probably be there | 20:08 |
| jeblair | i'll be going. | 20:29 |
| jeblair | i'd like to use the time to have design discussions and future-looking planning (eg, containers, etc). so if folks can make it, i think it'll be useful. we can try to conference in remote folks, but the more in person, the better i think. | 20:29 |
| ianw | jeblair: ok, the original idea was to test bits like argparse in here so before the rpc layer. but i can also *add* more end-to-end testing too | 20:30 |
| ianw | jeblair: also, i don't know if you'd accept it before a v3 branch, but it's a minor pet peeve of mine with required --args ... they should just be positional. i'll change that in the client if others feel the same :) | 20:35 |
| *** jasondot_ has joined #zuul | 20:39 | |
| *** jasondotstar has quit IRC | 20:43 | |
| jeblair | ianw: i like testing the argparse stuff. i just don't think it's necessary to encode the rpc format into that test. if we combine it with the other tests, we'll be testing exactly what needs to be tested and no more | 20:43 |
| Shrews | i prefer just the opposite, actually. i find it easier to remember the option names than the order of arguments | 20:44 |
| *** threestrands has joined #zuul | 20:47 | |
| *** threestrands has quit IRC | 20:47 | |
| *** threestrands has joined #zuul | 20:47 | |
| Shrews | i can never remember the order for 'ln -s' | 20:47 |
| jeblair | it's always the other order | 20:47 |
| Shrews | yes. always | 20:48 |
| *** neatherweb has quit IRC | 21:31 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Combine tox-siblings and tox roles https://review.openstack.org/523950 | 21:35 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Remove tox-siblings role stub https://review.openstack.org/523996 | 21:43 |
| openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: Normalize daemon process handling https://review.openstack.org/517381 | 21:47 |
| *** dkranz has quit IRC | 21:51 | |
| rcarrillocruz | i'd love to go to PTG | 22:02 |
| rcarrillocruz | will ask, not sure if i can justify business due to tooling we use in our CI etc | 22:02 |
| rcarrillocruz | at least is not US/APJ, so should be way cheaper, maybe i can make it thru to my manager | 22:03 |
| rcarrillocruz | cos i'd love to talk about ansible driver | 22:03 |
| *** jkilpatr has quit IRC | 22:05 | |
| *** hashar has quit IRC | 22:07 | |
| *** hashar has joined #zuul | 22:11 | |
| rcarrillocruz | tristanC: when you get around, what issues permissions for on https://review.openstack.org/#/c/522420/4/doc/source/admin/drivers/github.rst | 22:17 |
| rcarrillocruz | unless there's some 'Closes-Bug blah' integration don't see why is that needed ? | 22:17 |
| jeblair | rcarrillocruz, tristanC: good question -- we don't have that set up for the openstack-infra app. of course, we haven't exercised everything there yet. | 22:21 |
| *** jkilpatr has joined #zuul | 22:29 | |
| *** hashar has quit IRC | 22:29 | |
| rcarrillocruz | OH | 22:39 |
| rcarrillocruz | zuul consuming username from nodepool landed | 22:39 |
| rcarrillocruz | neat! | 22:40 |
| pabelanger | rcarrillocruz: \o/ | 22:41 |
| tristanC | rcarrillocruz: jeblair: i'm not entirely sure, but otherwise github wouldn't send pullrequest_comment, when adding issues it suddenly started to send issue_comment | 23:07 |
| rcarrillocruz | LOL | 23:07 |
| rcarrillocruz | SO | 23:07 |
| rcarrillocruz | you just hinted me why my 'recheck' doesn't work | 23:08 |
| rcarrillocruz | haha | 23:08 |
| rcarrillocruz | let me add that! | 23:08 |
| rcarrillocruz | yay to bogus GH permissions, facepalm... | 23:08 |
| tristanC | rcarrillocruz: i've a test pr with about 50 'recheck' comment... | 23:08 |
| rcarrillocruz | lol | 23:09 |
| rcarrillocruz | i gave up, never got my zuul app to send those events | 23:09 |
| * rcarrillocruz trying | 23:09 | |
| tristanC | rcarrillocruz: you have to 'update' the app permission on your project, at some point i was also removing and re-installing the app just to be sure | 23:10 |
| jeblair | pull requests are internally implemented as issues in github. that may explain the connection. however, since there are *also* pull request permissions, it's still a little surprising. | 23:11 |
| * jlk perks up | 23:12 | |
| jlk | oh oh | 23:12 |
| jlk | hehe this is a fun one | 23:12 |
| jlk | PullRequests are Issues, but Issues aren't Pull Requests | 23:13 |
| rcarrillocruz | 2017-11-29 23:13:01,968 DEBUG zuul.GithubWebhookListener: X-Github-Event: issue_comment | 23:13 |
| jlk | a comment on an issue and a comment on a pull request are basically the same thing, and the event is an issue_comment event. | 23:13 |
| rcarrillocruz | TA-DA | 23:13 |
| rcarrillocruz | thx tristanC :-) | 23:13 |
| clarkb | not sure if it is still the case but to close pull requests via the api you had to use the close issue api | 23:13 |
| clarkb | so ya lots of tangled relationship there | 23:14 |
| tristanC | rcarrillocruz: you're welcome | 23:14 |
| jlk | However, when looking at the code review tab, and commenting in-line in the code, that's a 'pull_request_review_comment' event. | 23:14 |
| tristanC | rcarrillocruz: would you mind checking if your permission matched the one i documented 522420? | 23:15 |
| jlk | not to be confused with a 'pull_request_review' event, which covers the newish "Reviews" that people can do on pull requests. | 23:15 |
| rcarrillocruz | yeah, as a matter of fact | 23:15 |
| rcarrillocruz | i jsut created another dummy org | 23:15 |
| jlk | just think of how everything in OpenStack (except for Swift) grew out of Nova, and what kind of gremlins that's left all throughout | 23:15 |
| rcarrillocruz | and creating another gh app | 23:15 |
| rcarrillocruz | in order to simulate a third party CI driven by GitHub | 23:15 |
| rcarrillocruz | will get back to you if your permissions are golden (it should, cos the only odd thing i spotted was the issues thing, and i never got recheck working , soo...) | 23:16 |
| dmsimard | Ansible 2.4.2 is out | 23:32 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Move to dictionary list of projects zuul._projects (take 2) https://review.openstack.org/518815 | 23:56 |
| mordred | dmsimard: woot. we should get around to upgrading | 23:57 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: web: add /{tenant}/jobs route https://review.openstack.org/503270 | 23:59 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: web: add /{tenant}/builds route https://review.openstack.org/466561 | 23:59 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: web: make console-stream tenant scoped https://review.openstack.org/505452 | 23:59 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: web: add /{source}/{project}.pub route https://review.openstack.org/502530 | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!