| *** _erlon_ has quit IRC | 00:06 | |
| *** rfolco has joined #zuul | 00:15 | |
| *** rfolco has quit IRC | 00:41 | |
| *** rfolco has joined #zuul | 01:12 | |
| *** hamalq_ has quit IRC | 01:21 | |
| *** Goneri has quit IRC | 01:23 | |
| *** rfolco has quit IRC | 02:15 | |
| *** piotrowskim has quit IRC | 02:26 | |
| *** freenzyfriday has joined #zuul | 03:10 | |
| *** freenzyfriday has quit IRC | 03:15 | |
| *** cloudnull has quit IRC | 03:34 | |
| *** cloudnull has joined #zuul | 03:34 | |
| *** bhavikdbavishi has joined #zuul | 03:37 | |
| *** bhavikdbavishi1 has joined #zuul | 03:48 | |
| *** bhavikdbavishi has quit IRC | 03:49 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 03:49 | |
| *** cloudnull4 has joined #zuul | 03:54 | |
| *** cloudnull has quit IRC | 03:55 | |
| *** cloudnull4 is now known as cloudnull | 03:55 | |
| *** bhavikdbavishi has quit IRC | 04:12 | |
| *** bhavikdbavishi has joined #zuul | 04:13 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #zuul | 04:33 | |
| bolg | clarkb: yes, i aboned the other one | 05:23 |
|---|---|---|
| *** reiterative has quit IRC | 05:42 | |
| *** reiterative has joined #zuul | 05:42 | |
| *** bhagyashri is now known as bhagyashris | 06:02 | |
| *** bhavikdbavishi1 has joined #zuul | 06:03 | |
| *** bhavikdbavishi has quit IRC | 06:05 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 06:05 | |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Separate connection registries in tests https://review.opendev.org/712958 | 06:21 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Separate connection registries in tests https://review.opendev.org/712958 | 06:23 |
| *** mach1na has joined #zuul | 06:25 | |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Prepare Zookeeper for scale-out scheduler https://review.opendev.org/717269 | 06:26 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Mandatory Zookeeper connection for ZuulWeb in tests https://review.opendev.org/721254 | 06:29 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Driver event ingestion https://review.opendev.org/717299 | 06:29 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect merger to Zookeeper https://review.opendev.org/716221 | 06:29 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect fingergw to Zookeeper https://review.opendev.org/716875 | 06:29 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect executor to Zookeeper https://review.opendev.org/716262 | 06:29 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Improve typings in context of 744416 https://review.opendev.org/753578 | 06:29 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Merge Zookeeper connection methods and specialize exceptions https://review.opendev.org/754360 | 06:29 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Switch to using zookeeper instead of gearman for jobs https://review.opendev.org/744416 | 06:35 |
| *** jhesketh has quit IRC | 06:35 | |
| *** jcapitao has joined #zuul | 06:44 | |
| *** stevthedev has quit IRC | 06:54 | |
| *** hamalq has joined #zuul | 06:57 | |
| *** hashar has joined #zuul | 06:59 | |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Prepare Zookeeper for scale-out scheduler https://review.opendev.org/717269 | 07:04 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Mandatory Zookeeper connection for ZuulWeb in tests https://review.opendev.org/721254 | 07:04 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Driver event ingestion https://review.opendev.org/717299 | 07:04 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect merger to Zookeeper https://review.opendev.org/716221 | 07:04 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect fingergw to Zookeeper https://review.opendev.org/716875 | 07:04 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect executor to Zookeeper https://review.opendev.org/716262 | 07:04 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Improve typings in context of 744416 https://review.opendev.org/753578 | 07:04 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Merge Zookeeper connection methods and specialize exceptions https://review.opendev.org/754360 | 07:04 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Switch to using zookeeper instead of gearman for jobs https://review.opendev.org/744416 | 07:04 |
| *** jhesketh has joined #zuul | 07:10 | |
| fbo | fungi: hi, no for the moment all have been merged. And yes let me provide a patch to remove that no longer needed warning. Thanks for the head up | 07:25 |
| openstackgerrit | Fabien Boucher proposed zuul/zuul master: gitlab - doc - remove not feature complete notice https://review.opendev.org/755435 | 07:30 |
| *** bhavikdbavishi has quit IRC | 07:30 | |
| *** tosky has joined #zuul | 07:39 | |
| *** jpena|off is now known as jpena | 07:56 | |
| *** stevthedev has joined #zuul | 08:07 | |
| *** bhavikdbavishi has joined #zuul | 08:12 | |
| *** bhavikdbavishi1 has joined #zuul | 08:15 | |
| *** bhavikdbavishi has quit IRC | 08:17 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 08:17 | |
| *** hamalq has quit IRC | 08:18 | |
| *** avass has joined #zuul | 08:23 | |
| openstackgerrit | Andy Ladjadj proposed zuul/nodepool master: [provider][static] add external_id field like other provider https://review.opendev.org/755443 | 08:27 |
| *** persia has quit IRC | 08:32 | |
| *** persia has joined #zuul | 08:32 | |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Add zuul-client-zuul-functional job https://review.opendev.org/753096 | 08:34 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Test zuul-client encrypt subcommand https://review.opendev.org/754103 | 08:38 |
| *** stevthedev has quit IRC | 08:52 | |
| openstackgerrit | Merged zuul/zuul-client master: Add cross testing with Zuul https://review.opendev.org/751264 | 08:53 |
| *** hamalq has joined #zuul | 09:00 | |
| *** hamalq has quit IRC | 09:04 | |
| *** stevthedev has joined #zuul | 09:10 | |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Separate connection registries in tests https://review.opendev.org/712958 | 09:43 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Prepare Zookeeper for scale-out scheduler https://review.opendev.org/717269 | 09:43 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Mandatory Zookeeper connection for ZuulWeb in tests https://review.opendev.org/721254 | 09:48 |
| *** hamalq has joined #zuul | 10:05 | |
| *** hamalq has quit IRC | 10:09 | |
| *** jcapitao has quit IRC | 10:13 | |
| *** holser has quit IRC | 10:14 | |
| *** holser has joined #zuul | 10:16 | |
| *** bhavikdbavishi has quit IRC | 10:32 | |
| *** mach1na has quit IRC | 10:56 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Warn user when dynamic layout ignores zuul config https://review.opendev.org/720249 | 11:10 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Make reporting asynchronous https://review.opendev.org/691253 | 11:24 |
| *** bhavikdbavishi has joined #zuul | 11:25 | |
| *** jpena is now known as jpena|lunch | 11:27 | |
| mhu | tobiash, the zuul-client encrypt patch chain should be ready for merging: https://review.opendev.org/#/q/topic:zuul-client_encrypt+(status:open) | 11:27 |
| *** mach1na has joined #zuul | 11:29 | |
| tobiash | mhu: is the testing strategy already fleshed out? (regarding your mail) | 11:29 |
| mhu | tobiash, clarkb's suggestion is the simplest way to go, but it adds the whole testing framework to the zuul package | 11:30 |
| mhu | until there's more feedback and a consensus is reached, I guess we can work with a "decoupled" testing approach (add the feature in zuul-client and submit a test in zuul) | 11:31 |
| mhu | especially for the time being as I am the only contributor to the lib so far :) | 11:32 |
| tobiash | k | 11:32 |
| tobiash | I wonder if it makes sense to leverage the quick-start job to get a running zuul and run the client tests against that | 11:34 |
| *** mach1na has quit IRC | 11:34 | |
| tobiash | mhu: commented on 754103 | 11:36 |
| *** rfolco has joined #zuul | 11:57 | |
| *** rlandy has joined #zuul | 11:58 | |
| *** rlandy is now known as rlandy|rover | 11:59 | |
| *** mach1na has joined #zuul | 12:00 | |
| *** hamalq has joined #zuul | 12:06 | |
| *** hamalq has quit IRC | 12:10 | |
| *** mach1na has quit IRC | 12:15 | |
| *** mach1na has joined #zuul | 12:19 | |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Configure redux for development https://review.opendev.org/755499 | 12:22 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Avoid state mutations in build reducers https://review.opendev.org/755500 | 12:22 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Improve fetch build actions and state-to-props handling on BuildPage https://review.opendev.org/755501 | 12:22 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Don't overwrite builds when retrieving the buildset https://review.opendev.org/755502 | 12:22 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Only request the buildset if it's not already available https://review.opendev.org/755503 | 12:22 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Add redux integration tests for build actions https://review.opendev.org/755504 | 12:22 |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: Validate ensure-docker works https://review.opendev.org/755505 | 12:26 |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: Update ensure-docker for new releases https://review.opendev.org/752630 | 12:28 |
| openstackgerrit | Merged zuul/zuul master: javascript: publish artifact tarball https://review.opendev.org/754245 | 12:29 |
| *** jpena|lunch is now known as jpena | 12:29 | |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Configure redux for development https://review.opendev.org/755499 | 12:38 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Avoid state mutations in build reducers https://review.opendev.org/755500 | 12:38 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Improve fetch build actions and state-to-props handling on BuildPage https://review.opendev.org/755501 | 12:38 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Don't overwrite builds when retrieving the buildset https://review.opendev.org/755502 | 12:38 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Only request the buildset if it's not already available https://review.opendev.org/755503 | 12:38 |
| openstackgerrit | Felix Edel proposed zuul/zuul master: Add redux integration tests for build actions https://review.opendev.org/755504 | 12:38 |
| felixedel | zuul-maint: I have a bunch of redux related changes that are ready for review https://review.opendev.org/#/q/topic:redux+(status:open+OR+status:merged)+AND+project:zuul/zuul :) They add some useful extensions for redux development, fix some bugs in the build actions and add some tests | 12:41 |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: Validate ensure-docker works https://review.opendev.org/755505 | 12:48 |
| *** Goneri has joined #zuul | 12:59 | |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Prepare Zookeeper for scale-out scheduler https://review.opendev.org/717269 | 13:01 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Mandatory Zookeeper connection for ZuulWeb in tests https://review.opendev.org/721254 | 13:01 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Driver event ingestion https://review.opendev.org/717299 | 13:01 |
| lyr | Hi there | 13:13 |
| lyr | I'm looking for some troubleshooting doc, our jobs aren't starting, and I can't find any hints | 13:13 |
| openstackgerrit | Merged zuul/zuul master: Add zuul-client-zuul-functional job https://review.opendev.org/753096 | 13:14 |
| corvus | lyr: you can check the scheduler log | 13:20 |
| lyr | nothing much in the log corvus | 13:21 |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: Validate ensure-docker works https://review.opendev.org/755505 | 13:22 |
| corvus | lyr: is it logging at debug level? | 13:22 |
| lyr | yes | 13:22 |
| corvus | lyr: then it should have the complete decision making process for running jobs. if there's nothing in there, then you may want to start with whether it's actually getting any events. | 13:24 |
| lyr | a colleague re-checked & re-gated a pair of jobs after a zuul restart, they do appear in the status UI as queued, but nothing happens then | 13:24 |
| corvus | lyr: if they are in the ui, there should definitely be entries in the scheduler log about them | 13:25 |
| lyr | here's the UI screenshot https://i.imgur.com/4mZt9Nx.png | 13:25 |
| lyr | here's a bit of log grep'ed by org name https://i.imgur.com/MY7zx8x.png | 13:28 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect merger to Zookeeper https://review.opendev.org/716221 | 13:32 |
| corvus | lyr: closer to when the items were added to the queue, you should see log entries about requesting nodes for the job, or, once they are ready, submitting the job to the executors. zuul is likely either waiting for nodes, or there are no executors on-line to run the job. | 13:33 |
| openstackgerrit | Merged zuul/zuul master: web: consistent font sizes on console page and PF4 for task summary https://review.opendev.org/752383 | 13:36 |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: Validate ensure-docker works https://review.opendev.org/755505 | 13:41 |
| lyr | hmm running "nodepool list" all nodes are in "deleting" state. Before restarting they weren't in this state, the log if a bit far scrolling, but I remember it was mixed "used" and "available", or synonym words | 13:42 |
| corvus | lyr: yeah, 'used' means zuul has claimed it and has handed it off to the executor to run the job. so if there are zero used nodes, that's a clue it's waiting for nodes. in the scheduler log, you should see the nodepool request id it submitted for the job, and you can grep for that in the nodepool launcher logs. you can also run 'nodepool request-list' and see the list of requests with ids and their | 13:44 |
| corvus | statuses. | 13:44 |
| lyr | request-list is empty | 13:45 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect fingergw to Zookeeper https://review.opendev.org/716875 | 13:49 |
| fungi | zuul-maint: we have a bit of a quandry with ensure-twine in opendev and are trying to work out the best way to solve it | 13:53 |
| fungi | apparently twine depends on cmarkgfm which doesn't publish any wheels for python 3.8 | 13:54 |
| fungi | so since the docker images switched to being python 3.8 based we've been unable to run the ensure-twine role on our executors | 13:54 |
| tobiash | so should we switch back to 3.7 for now? | 13:55 |
| fungi | as the executor images don't come with a compiler toolchain sufficient to build the cmarkgfm sdist | 13:55 |
| zbr | fungi: i am curious about details, as I seen a lot of duplicated effort (or extra complexity) for stuff like ensure-(pip|setuptools|virtualenv|twine|tox) stuff. | 13:55 |
| tobiash | is this running on the executor? | 13:55 |
| fungi | yes | 13:55 |
| fungi | on the executor | 13:55 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Revert "Update images to use python 3.8" https://review.opendev.org/755518 | 13:57 |
| corvus | lyr: then i think the next step would be to continue to look in the scheduler debug logs to find out why zuul enqueued a change but there is no node request for it. | 13:57 |
| tobiash | the revert in case there is a decision to switch back to 3.7 ^ | 13:57 |
| fungi | i'm not too excited about the prospect of holding zuul back from using current python interpreters in its container images, but yes that would be one option i guess | 13:58 |
| fungi | as would trying to roll opendev's executors back to older docker images | 13:58 |
| fungi | but moreso i'd like to figure out a way forward to be able to have python 3.8 and still run twine | 13:59 |
| corvus | fungi: it's unfortunate that twine of all things has this problem. at one point we uploaded without twine to avoid dependencies, but i guess the protocol got too complicated? | 13:59 |
| lyr | corvus, I got a "clean" log here https://paste.garrigue.re/?e5b94f28126af34e#aeHRhamT3pqxowpaxa4XLs2y6cy5dLEyeuH5y7i7QB4, looking at it atm, if you've some old timer ctrl-f to check this, be my guest :D | 13:59 |
| corvus | fungi, tobiash: we could add twine (and/or it's dependencies) to the executor image? | 14:00 |
| corvus | it meets the 'used by zuul-jobs' requirement | 14:00 |
| tobiash | that's another option (if we install and uninstall the toolchain) | 14:00 |
| fungi | that would be one way... what sort of package management is available in the image build process? debian/sid? | 14:01 |
| tobiash | I guess that's a little bit more complicated | 14:01 |
| tobiash | we could also do the revert, work on twine in executor and then unrevert again | 14:01 |
| tobiash | if there is a time constraint to fix it | 14:01 |
| corvus | tobiash: that seems prudent | 14:01 |
| fungi | also it's a very python-centric tool, and while zuul is written in python it seems slightly weird to bake something like twine into the executor images | 14:01 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul-client master: Add Dockerfile https://review.opendev.org/755519 | 14:02 |
| corvus | lyr: see if you can find node request 100-0001662935 in the nodepool launcher logs | 14:02 |
| fungi | i also wonder if there won't be other things in time that we want to pip install in an executor workspace which also need a compiler | 14:03 |
| fungi | but working out the additional c libs for different packages is still a sticking point i suppose | 14:04 |
| ttx | from a user perspective, +1 on unblocking us with a quick revert while the correct solution is devised | 14:04 |
| corvus | fungi: 755518 awaits your +3 | 14:05 |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Connect executor to Zookeeper https://review.opendev.org/716262 | 14:05 |
| lyr | corvus: here it is https://paste.garrigue.re/?8292ebbebbb8b062#3afzH9y1P7nz51mbdcpkNQVr4Cjwe6cg484LdD2hXyRL | 14:05 |
| corvus | lyr: i mean look for that in the nodepool launcher logs | 14:06 |
| lyr | corvus: you mean the infrastructure providing the node ? (openstack in our case) | 14:06 |
| *** hamalq has joined #zuul | 14:07 | |
| corvus | lyr: i mean the zuul component called "nodepool-launcher" | 14:07 |
| corvus | it's a separate process that interfaces between zuul and openstack | 14:07 |
| fungi | corvus: tobiash: ttx: i've approved 755518, but still would like to continue brainstorming how we might support things like twine on the executor (or do we need to start using it on job nodes where we can install things as root?) | 14:08 |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: Validate ensure-docker works https://review.opendev.org/755505 | 14:10 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Test zuul-client encrypt subcommand https://review.opendev.org/754103 | 14:11 |
| ttx | fungi: is there a specific reason (like access to a secret) why that part of the publish-pypi job runs on the executor? | 14:11 |
| *** hamalq has quit IRC | 14:11 | |
| fungi | i'll have to go back over our job to say for certain | 14:12 |
| corvus | yes, it has access to our publication secret and we don't want it to run on a tainted node | 14:12 |
| fungi | thanks, i suspected that might be the case | 14:13 |
| fungi | but couldn't remember for sure | 14:13 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: Test zuul-client encrypt subcommand https://review.opendev.org/754103 | 14:13 |
| fungi | so if we want to recommend python package uploads happen from an executor workspace then the options are to write a twine replacement which doesn't have binary dependencies, add twine and its deps to the executor images, add a compiler toolchain and common c headers to the images... what other options? | 14:15 |
| corvus | that's all i got | 14:15 |
| fungi | oh, create our own manylinux1 wheels for the binary deps of twine | 14:15 |
| corvus | oh -- ask twine to pls have wheels for its deps or reduce its deps | 14:15 |
| corvus | ya that :) | 14:15 |
| fungi | anyway, i'll start on a bug report for cmarkgfm if there isn't one already, requesting a py38 wheel | 14:16 |
| fungi | as soon as i get some more coffee | 14:17 |
| *** bhavikdbavishi has quit IRC | 14:21 | |
| *** hashar has quit IRC | 14:21 | |
| lyr | Ok, fixed, thanks corvus | 14:21 |
| lyr | I don't know why it was stuck to begin with | 14:21 |
| lyr | But nuking existing nodes & restarting node launcher kicked it on track again | 14:22 |
| lyr | Thnks for your help | 14:23 |
| corvus | lyr: np, sorry we didn't get to the root cause | 14:24 |
| zbr | zuul-maint: https://review.opendev.org/#/c/755505/ -- improves testing of ensure-docker, needed for follow-up. | 14:24 |
| lyr | Well, we got this zuul around, but not the guy who pushed for it anymore | 14:24 |
| lyr | It's not maintained like it should be, no update, no monitoring, ... | 14:25 |
| lyr | I'm just praying it holds up until I can manage it properly. Next year maybe... | 14:25 |
| corvus | lyr: we're all lazy sysadmins, so we're all motivated to try to make it run by itself :) | 14:26 |
| zbr | lyr: be sure you pray well. zuul is very reliable but has a maintenance cost that small shops may not be able (or willing) to pay for. | 14:30 |
| zbr | corvus: can you please help with ensure-dockre ^ above? My test uncovers that ensure-docker is partially broken on two platforms, some i do not care much about. | 14:32 |
| zbr | mainly what i did was some kind of xfail implementaito | 14:32 |
| lyr | zbr yeah, we were meant to get bigger, but growth is stalled atm | 14:39 |
| zbr | lyr: you may want to check https://vexxhost.com/solutions/managed-zuul/ -- it may be a viable option if you lack internal knowledge on how to manage it. I am not affiliated but it would not hurt to get some info from there. | 14:41 |
| fungi | the folks who run that one are super nice (i'm also not affiliated with them, fwiw, but some of them are involved in the project) | 14:44 |
| fungi | depends a lot on what you're trying to use zuul for though... if you have an environment where the components need to live in your own lab network or something then relying on a third party's external saas likely won't work for you | 14:45 |
| zbr | fungi: do you know a way to trigger some kind of zuul warning from inside an ansible task run by zuul? like passing the job but with a visible warning. | 14:48 |
| zbr | i am trying to make the xfail more visible and i suppose we may need the some in the future. | 14:49 |
| zbr | it is unreasonable to prevent a fix for platform A from being merged because platform B is already broken but we didn't know about it. Also relying on user to fix existing unrelated bugs goes into the same category. | 14:50 |
| zbr | as number of platforms and roles grew the change of such conflicts grew | 14:51 |
| fungi | zbr: "visible warning" visible where? | 14:51 |
| fungi | visible in the job log? in the webui's chrome? in comments on a proposed change? | 14:52 |
| zbr | somewhere on zuul message posted to gerrit i think, other places are not guaranteed to be checked by reviewers. | 14:52 |
| openstackgerrit | Merged zuul/zuul master: gitlab - doc - remove not feature complete notice https://review.opendev.org/755435 | 14:53 |
| fungi | would there be a particular line number for the diff which should get highlighted as responsible for the warning? | 14:53 |
| avass | zbr: commented :) | 14:54 |
| zbr | the issue that that the source of the error may not be part of the edited code | 14:54 |
| zbr | gh-actions is able to add comments to any piece of code, even if is not part of the changeset, but gerrit does not allow that. | 14:54 |
| zbr | we may try to use comment on commit message, as that always exists. | 14:54 |
| corvus | file comments are an option | 14:55 |
| *** sduthil has joined #zuul | 14:56 | |
| fungi | i think zuul already tries to address that by adding a review comment stating that there were references in the job output which indicated lines in other files not included in the diff, and to look at the build log for details | 14:56 |
| zbr | ideally we should have a WARN / orange result code, so it would be obvious in UI that is just a partial/impure success. | 14:56 |
| zbr | this should not be confused with non-voting, which is always ignored | 14:56 |
| fungi | i'm unconvinced that there is such a thing s "partial success" | 14:56 |
| fungi | there's either a condition which should prevent a change from merging, or it's safe to merge. trying to use build reports on proposed changes as a signal for code problems which aren't part of the merge decision seems likely to cause more problems than it solves | 14:58 |
| fungi | identifying those issues is probably better done post-merge, say in periodic jobs | 14:58 |
| AJaeger | clarkb: did you get an answer on 747732? Is it safe to merge that? | 14:58 |
| zbr | fungi: take a look at last task on https://review.opendev.org/#/c/755505/5/test-playbooks/ensure-docker.yaml | 14:59 |
| clarkb | AJaeger: I haven't seen one. I guess we go with it since testing was happy? | 15:00 |
| zbr | how can I tell (display) to the average user that there is a known issue. It is unlikely that reviewers will check full logs of all jobs and spot such issue. | 15:01 |
| zbr | if we remind them that a job passed, but with some warnings, they would be reminded about it and more likely to try to address it | 15:01 |
| avass | zbr: --rm should only remove the container and not the image you're using | 15:02 |
| AJaeger | clarkb: if somebody is around in case testing was not sufficient ;) | 15:02 |
| zbr | avass: ahh, right. i will fix. other comments? | 15:03 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul-client master: Add Dockerfile https://review.opendev.org/755519 | 15:04 |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: Validate ensure-docker works https://review.opendev.org/755505 | 15:07 |
| fungi | so circling back to the ensure-twine problem, looks like there's been a request open since january to add py38 wheels for cmarkgfm: https://github.com/theacodes/cmarkgfm/issues/22 | 15:15 |
| openstackgerrit | Merged zuul/zuul master: Revert "Update images to use python 3.8" https://review.opendev.org/755518 | 15:21 |
| *** mach1na has quit IRC | 15:21 | |
| avass | zbr: I don't think so, no | 15:24 |
| AJaeger | clarkb: shall I approve 747732 now? | 15:26 |
| AJaeger | or do you want to? | 15:27 |
| clarkb | AJaeger: heh I think that number is wrong now that I've pulled it up in gerrit to dobule check | 15:27 |
| AJaeger | https://review.opendev.org/#/c/742732/ | 15:27 |
| clarkb | AJaeger: I'm hoping to spend my morning bootstrapping review-test.o.o so will probably be distracted. | 15:27 |
| AJaeger | clarkb: indeed, one wrong digit | 15:27 |
| clarkb | I think you can approve it and if things go poorly I should be able to help? maybe corvus understands the ansible there better than I do? | 15:27 |
| AJaeger | approved now... | 15:28 |
| avass | I can take a look at it as well. I got some time to spare anyway :) | 15:28 |
| AJaeger | avass: Ok, let me remove the +A again to give you time to review | 15:29 |
| AJaeger | avass: see clarkb's comment on http://eavesdrop.openstack.org/irclogs/%23zuul/%23zuul.2020-09-30.log.html#t2020-09-30T15:25:08 | 15:30 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul-client master: Add Dockerfile https://review.opendev.org/755519 | 15:31 |
| openstackgerrit | Matthieu Huin proposed zuul/zuul master: REST API: improve tenant scoping of autohold, authorizations https://review.opendev.org/728118 | 15:32 |
| mhu | Hello zuul-maint, I'm adding a container image for the zuul-client CLI, what is the process to get docker credentials? see https://review.opendev.org/#/c/755519/ | 15:34 |
| fungi | mhu: someone who has the docker creds for the zuul namespace will need to encrypt a secret and update that change or add a followup change with it | 15:36 |
| mhu | fungi, okay, do I need to notify anyone in particular? | 15:38 |
| *** hashar has joined #zuul | 15:40 | |
| avass | AJaeger: yeah I read that as well, I guess it's up for debate what the documentation means with 'execute role dependencies' since upload-logs-base is only used as a library in this case | 15:41 |
| avass | but it shouldn't be too hard to test it | 15:41 |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: Validate ensure-docker works https://review.opendev.org/755505 | 15:41 |
| fungi | mhu: looks like corvus added the equivalent secret to the zuul/zuul repo, so maybe he'd be up for doing that again for zuul/zuul-client, but looks like the secret is currently held by the opendev sysadmins in our credential list too so i could add it if he's busy | 15:41 |
| avass | AJaeger, clarkb: upload-logs-s3 is tested with include_role https://opendev.org/zuul/zuul-jobs/src/branch/master/test-playbooks/upload-logs-s3.yaml#L50 | 15:42 |
| mhu | fungi, thanks, there's no rush obviously, as long as it happens at some point | 15:43 |
| avass | AJaeger, clarkb: so I guess we already have a testcase for that if I understand the problem correctly :) | 15:43 |
| AJaeger | avass: thanks for confirming - do want to +2A? | 15:44 |
| avass | wf+1? | 15:45 |
| avass | sure :) | 15:45 |
| AJaeger | yep | 15:45 |
| AJaeger | thanks | 15:45 |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: Update ensure-docker for new releases https://review.opendev.org/752630 | 16:00 |
| openstackgerrit | Merged zuul/zuul-jobs master: Merge upload logs modules into common role https://review.opendev.org/742732 | 16:00 |
| *** hamalq has joined #zuul | 16:00 | |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: Update ensure-docker for new releases https://review.opendev.org/752630 | 16:02 |
| *** hamalq has quit IRC | 16:04 | |
| *** hamalq has joined #zuul | 16:04 | |
| *** bhavikdbavishi has joined #zuul | 16:06 | |
| *** hamalq has quit IRC | 16:07 | |
| *** hamalq has joined #zuul | 16:08 | |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: tutorial: Rework quick-start and prepare for other tutorials https://review.opendev.org/732066 | 16:19 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: tutorial: Add "gate your first patch" https://review.opendev.org/732067 | 16:19 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: tutorial: Add "Use zuul jobs" https://review.opendev.org/732068 | 16:19 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: tutorial: Add "gate pipeline" https://review.opendev.org/732069 | 16:19 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: tutorial: Add "job secrets" https://review.opendev.org/732070 | 16:19 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: tutorial: Add "job dependencies" https://review.opendev.org/732071 | 16:19 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: Rename quick-start to zuul-tutorial-quick-start https://review.opendev.org/737656 | 16:19 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: [DNM] TEST run zuul tutorials to test stream+callback (+ zuul-jobs change) https://review.opendev.org/735477 | 16:19 |
| openstackgerrit | Guillaume Chauvel proposed zuul/zuul master: [DNM] Test: run multiple tutorials ('job dependencies' 2 times) https://review.opendev.org/741558 | 16:19 |
| *** mattd01 has joined #zuul | 17:01 | |
| *** mattd01 has left #zuul | 17:03 | |
| *** mattd01 has joined #zuul | 17:03 | |
| *** jpena is now known as jpena|off | 17:05 | |
| *** mattd01 has quit IRC | 17:08 | |
| *** bhavikdbavishi1 has joined #zuul | 17:26 | |
| *** bhavikdbavishi has quit IRC | 17:28 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 17:28 | |
| *** bhavikdbavishi has quit IRC | 17:30 | |
| *** bhavikdbavishi has joined #zuul | 17:30 | |
| *** mattd01 has joined #zuul | 17:55 | |
| *** armstrongs has joined #zuul | 17:55 | |
| *** armstrongs has quit IRC | 17:58 | |
| *** saneax has quit IRC | 17:59 | |
| *** stevthedev has quit IRC | 18:00 | |
| *** stevthedev has joined #zuul | 18:47 | |
| *** iurygregory has quit IRC | 18:55 | |
| *** rlandy|rover is now known as rlandy|rover|mtg | 19:00 | |
| AJaeger | tobiash: did you see the comment/question on https://review.opendev.org/#/c/742736/10 ? | 19:19 |
| tobiash | AJaeger: thanks for the notice, I'll check tomorrow | 19:22 |
| *** rlandy|rover|mtg is now known as rlandy|rover | 19:30 | |
| *** bhavikdbavishi has quit IRC | 19:37 | |
| AJaeger | tobiash: sure - no urgency... enjoy the evening! | 19:38 |
| *** iurygregory has joined #zuul | 19:43 | |
| *** mattd01 has left #zuul | 19:47 | |
| *** sduthil has quit IRC | 20:21 | |
| openstackgerrit | Jan Kubovy proposed zuul/zuul master: Improve typings in context of 744416 https://review.opendev.org/753578 | 20:24 |
| *** hashar has quit IRC | 20:48 | |
| *** rfolco has quit IRC | 21:07 | |
| *** freenzyfriday has joined #zuul | 21:09 | |
| *** freenzyfriday has quit IRC | 21:14 | |
| *** holser has quit IRC | 21:25 | |
| *** mgoddard has quit IRC | 21:30 | |
| *** mgoddard has joined #zuul | 21:36 | |
| *** holser has joined #zuul | 21:42 | |
| ianw | doesn't look like https://github.com/theacodes/cmarkgfm has an CI for wheels | 21:46 |
| ianw | i.e. whatever has been uploaded for prior releases has probably been hand generated and put up there somehow | 21:47 |
| ianw | when you actually think about the implications of that ... well ... don't | 21:47 |
| fungi | yeah, it's on my to do list to ping thacodes on the 3.8 wheel request issue from january (and the 3.8 support meta addition pr from november) | 21:48 |
| ianw | perhaps i'm wrong, something something appveyor | 21:49 |
| fungi | though in #pypa-dev, dstufft pointed out that uploading an abi3 wheel would also prevent this coming up again when 3.9 releases in... days? | 21:49 |
| ianw | https://github.com/theacodes/cmarkgfm-wheels/commit/ddcda9c6a7910e193e4aefa40bedcdb72477ecc2 | 21:51 |
| ianw | Add Python 3.8 (and pray it works) | 21:51 |
| fungi | this also brought to the attention of the twine maintainers that the person in control of one of their dependencies is possibly missing in action | 21:52 |
| ianw | so the project has a separate wheels repo | 21:52 |
| fungi | that's an unusual choice | 21:53 |
| ianw | that relies on https://github.com/matthew-brett/multibuild which was either perfected several years ago, or is unmaintained | 21:53 |
| fungi | i'm going with perfected for all time | 21:54 |
| clarkb | I wish I could perfect things :) | 21:54 |
| fungi | that's easy, just stop working on them | 21:54 |
| *** freenzyfriday has joined #zuul | 21:55 | |
| ianw | if appveyor tried to build that, i'm stumpted as to where to look for it | 21:55 |
| ianw | https://ci.appveyor.com/project/theacodes/cmarkgfm-wheels/builds/30450845 i guess | 21:56 |
| clarkb | (and pray it works) | 21:57 |
| ianw | ok, i would say it looks like the 2.7 build failed, and then nothing else tried after that | 21:57 |
| clarkb | ya that is how I read that too | 21:57 |
| fungi | so stopped dead in its tracks by 2.7 eol? | 21:57 |
| ianw | i think more the job bit-rotted | 21:58 |
| ianw | basically i would not be holding one's breath for this | 21:59 |
| *** freenzyfriday has quit IRC | 21:59 | |
| ianw | the issue with offering to do the manylinux builds is that we can't offer windows and osx like these other services | 22:12 |
| ianw | unfortunately I also just noticed the maintainer's github status is "Mental health leave - almost certainly not going to respond" | 22:28 |
| *** rlandy|rover has quit IRC | 22:34 | |
| openstackgerrit | Pierre-Louis Bonicoli proposed zuul/zuul master: smtp reporter: authenticate with client certificate https://review.opendev.org/755668 | 22:40 |
| *** tosky has quit IRC | 22:45 | |
| clarkb | ianw: ya there was discussion in pypa-dev if you want to catch up on what was said there | 22:46 |
| clarkb | (sorry been distracted by gerrit things) | 22:46 |
| ianw | Pygments-2.7.1 SecretStorage-3.1.2 bleach-3.2.1 certifi-2020.6.20 cffi-1.14.3 chardet-3.0.4 colorama-0.4.3 cryptography-3.1.1 docutils-0.16 idna-2.10 jeepney-0.4.3 keyring-21.4.0 packaging-20.4 pkginfo-1.5.0.1 pycparser-2.20 pyparsing-2.4.7 readme-renderer-26.0 requests-2.24.0 requests-toolbelt-0.9.1 rfc3986-1.4.0 six-1.15.0 tqdm-4.50.0 twine urllib3-1.25.10 webencodings-0.5.1 | 22:46 |
| ianw | my twine is not seeming to want to install cmarkgfm | 22:46 |
| *** freenzyfriday has joined #zuul | 22:46 | |
| clarkb | I think it comes from readme-renderer | 22:47 |
| ianw | Collecting cmarkgfm>=0.2.0; extra == "md" | 22:47 |
| clarkb | maybe that just updated? | 22:47 |
| clarkb | https://github.com/pypa/readme_renderer/blob/master/setup.py#L76-L78 ya | 22:48 |
| clarkb | how are we installing that then if it is an extra? | 22:48 |
| ianw | command: "{{ twine_python }} -m pip install twine!=1.12.0 readme_renderer[md]!=23.0 requests-toolbelt!=0.9.0 --user" | 22:49 |
| ianw | so i guess it's not so much a twine issue ... | 22:49 |
| clarkb | and we probably included that to do the package checking pre upload | 22:50 |
| clarkb | I wonder if we can make that a pre merge thing instead then do simple mode twine invocations for final upload | 22:50 |
| ianw | that was added to the role with https://review.opendev.org/#/c/613726/ | 22:51 |
| *** freenzyfriday has quit IRC | 22:51 | |
| clarkb | huh I wonder if we can just drop it then | 22:51 |
| fungi | that's a very good point | 22:52 |
| fungi | zbr probably has input when he's around again, since that was his addition | 22:52 |
| fungi | we don't actually need twine --check during the upload, just for proposed changes in check/gate | 22:53 |
| *** holser has quit IRC | 22:54 | |
| ianw | it seems that's two somewhat unrelated changes in 613726? | 22:54 |
| ianw | the != 23 has probably been resolved | 22:55 |
| ianw | this is also why i don't like linking to job failures but rather giving verbose explanations in changes | 22:56 |
| fungi | yes, including the error message is far more durable | 22:57 |
| ianw | anyway, it looks like the MD rendering thing is a warning anyway | 22:57 |
| ianw | if we wanted to maintain the status quo, we could leave it on by default, and put a flag to turn it off, noting that it pulls in binary dependencies | 22:58 |
| fungi | but i agree, it seems avoidable | 22:59 |
| clarkb | ya I think its to check that you can render the md at all | 23:00 |
| ianw | i can build a manylinux wheel using a bunch of hacked stuff cribbed from pypa, but i don't think i want to push on it | 23:00 |
| clarkb | and so the warning is basically saying " Ican't check this" | 23:00 |
| clarkb | we shouldn't need that during upload if we check it earlier | 23:00 |
| fungi | it's an excellent solution i hadn't spotted, and better than our other options in my opinion | 23:02 |
| ianw | i'll put a change and we can discuss | 23:02 |
| fungi | thanks!!! | 23:07 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Update Gerrit config for quickstart https://review.opendev.org/755669 | 23:10 |
| corvus | fungi, clarkb: ^ if you could speedy-approve that, it would aid my preparation for the summit talk i'm working on | 23:11 |
| clarkb | corvus: it has my +2 | 23:11 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: ensure-twine: remove readme_renderer https://review.opendev.org/755670 | 23:15 |
| fungi | corvus: that's much cleaner too | 23:18 |
| *** rfolco has joined #zuul | 23:19 | |
| corvus | is there a revert revert for 3.8? | 23:21 |
| clarkb | I/m not sure, the issue was largely handled by the time I started my day | 23:22 |
| corvus | i'll check/prep | 23:22 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Revert "Revert "Update images to use python 3.8"" https://review.opendev.org/755671 | 23:23 |
| corvus | that deps-on ianw's change | 23:23 |
| *** rfolco has quit IRC | 23:24 | |
| fungi | and i guess whatever ensure-twine role test we might have wouldn't take the executor image python interpreter version into account | 23:25 |
| *** rfolco has joined #zuul | 23:25 | |
| clarkb | fungi: and/or it runs with gcc installed | 23:27 |
| fungi | indeed | 23:27 |
| *** rfolco has quit IRC | 23:29 | |
| ianw | yeah, "must be executor safe" isn't really part of testing i guess | 23:29 |
| ianw | because it just runs on a node | 23:30 |
| openstackgerrit | Merged zuul/zuul-jobs master: ensure-twine: remove readme_renderer https://review.opendev.org/755670 | 23:32 |
| *** stevthedev has quit IRC | 23:54 | |
| *** Goneri has quit IRC | 23:55 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
