| === vrubiolo1 is now known as vrubiolo | |
| [13:18] <lucasmoura> Hey everyone, do we have a script called "sbuild-it" ? I am reviewing https://github.com/canonical/cloud-init/pull/407 and one of the step is to run: | |
| [13:18] <lucasmoura> sbuild-it ../out/cloud-init_20.2-45-g5f7825e2-0ubuntu1~18.04.1.dsc | |
| [13:19] <lucasmoura> Build I can't find this script in either uss-tableflip or qa-scripts | |
| [14:06] <Odd_Bloke> I'm not familiar with it. | |
| [14:18] <lucasmoura> Also, a question for a different issue, if the bug that I am doing the manual testing is not related to a launchpad bug, is there a convention to name that file ? | |
| [14:36] <rharper> lucasmoura: I think blackboxsw might have that; | |
| [14:38] <lucasmoura> rharper, ack | |
| [14:39] <rharper> lucasmoura: there's no convention yet; I suspect git log line <commish> replacing spaces with - | |
| [14:39] <lucasmoura> rharper, okay | |
| [14:39] <rharper> or, we can start using gh-<pr-number> | |
| [14:40] <rharper> Odd_Bloke: blackboxsw ^ ? for fixes without lp bugs to match in ubuntu-sru test-cases ? | |
| [14:44] <falcojr> is "cloud-init devel schema" supposed to work with jinja templates? | |
| [14:45] <falcojr> I get a 'Cloud config schema errors: format-l1.c1: File jinja.yaml needs to begin with "#cloud-config"' but our docs say the first line should be the jinja comment | |
| [14:48] <Odd_Bloke> That sounds like a gap we'll need to address as part of the schema work this cycle. | |
| [14:48] <rharper> falcojr you need both ##template: jinja\n#cloud-config\nruncmd: ['echo', 'Hi'] | |
| [14:49] <rharper> falcojr: oh I see; you want to schema check a template ... | |
| [14:49] <falcojr> thanks. I had another issue with my config so wanted to verify it with that before fully launching it | |
| [14:49] <falcojr> yeah | |
| [14:49] <rharper> that seems like a new use-case; as it's not*yet* fully formed | |
| [14:49] <rharper> until you render it | |
| [14:49] <rharper> the schema won't know what to do with {{ foo }} etc | |
| [14:50] <falcojr> right, makes sense | |
| [14:50] <rharper> let's file a bug/new-feature for that use-case; | |
| [14:52] <falcojr> I can do that (hopefully ;) ) | |
| [14:54] <Odd_Bloke> rharper: falcojr: As for naming, I think we've used git-<abbreviated hash>.txt before, but I don't have a strong preference. | |
| [14:56] <falcojr> Odd_Bloke: sorry, for naming what? | |
| [14:58] <falcojr> bug here: https://bugs.launchpad.net/cloud-init/+bug/1881925 | |
| [14:58] <ubot5> Ubuntu bug 1881925 in cloud-init "Schema validation fails on jinja template" [Undecided,New] | |
| [14:58] <Odd_Bloke> Oh sorry, that was lucasmoura not you. | |
| [14:59] <lucasmoura> Odd_Bloke, I have opened a PR with rharper naming suggestion. But I can rename it if the proposed convention should not be followed | |
| [14:59] <lucasmoura> Also, if my manual test references two distinct PRs, should I include both of them in the file name or just one ? In my PR I included just one, but I don't know if that is right | |
| [15:18] <rharper> cool | |
| [15:51] <blackboxsw> lucasmoura: thanks for the ping, I've added a PR which adds the sbuild-it script I use for package build and validation. it's smoser's script originally, but I think it makes sense in our uss-tableflip repo | |
| [15:53] <blackboxsw> lucasmoura: https://github.com/canonical/uss-tableflip/pull/52 | |
| [15:53] <blackboxsw> feel free to land and merge that into uss-tableflip if it looks good. | |
| [16:20] <Odd_Bloke> falcojr: I feel like we haven't reached consensus on my comment in https://github.com/canonical/cloud-init/pull/367/files#r434689535, gimme a shout if you want to jump on a call to get on the same page. | |
| [16:22] <falcojr> Odd_Bloke: now works for me. Standup room? | |
| [16:23] <Odd_Bloke> Sure, omw. | |
| [16:57] <blackboxsw> falcojr: nice schema devel gap/bug. the CLI tools wasn't full wired to the jinja part handling. so it's rudimentary. a fix would be helpful | |
| [17:00] <blackboxsw> landed https://github.com/canonical/uss-tableflip/pull/52 | |
| [17:00] <blackboxsw> so an update of uss-tableflip will expose scripts/sbuilt-it for package upload verification | |
| [17:04] <blackboxsw> also eoan, xenial and bionic upload PRs are all updated for review (so we can officially kickoff the SRU process with the 'final' bits). eoan: https://github.com/canonical/cloud-init/pull/409 bionic: https://github.com/canonical/cloud-init/pull/407 xenial: https://github.com/canonical/cloud-init/pull/406 | |
| [17:04] <blackboxsw> I think lucasmoura is on #407 already and dan on #409 | |
| === hjensas is now known as hjensas|afk | |
| [18:02] <blackboxsw> lucasmoura: sorry on the sbuild-it script, that should now be landed via the following merged pull request: https://github.com/canonical/uss-tableflip/pull/52 | |
| [18:03] <lucasmoura> blackboxsw, no worries. I have just used to build the bionic package and now I was able to build the package | |
| [18:03] <lucasmoura> Also, I just finished reviewing the bionic PR | |
| [18:21] <Odd_Bloke> blackboxsw: One typo in the eoan changelog, then I think it's good. Thanks! | |
| [19:32] <blackboxsw> thanks Odd_Bloke, just pushed the changelog fix | |
| [19:48] <Odd_Bloke> Thanks, looking now. | |
| [19:51] <Odd_Bloke> blackboxsw: What does sbuild-it do for us other than save you from substituting the correct dist in? | |
| [20:02] <blackboxsw> hrm, travis seems to have queued and not run the ubuntu/<series> branches for a while. | |
| [20:03] <blackboxsw> Odd_Bloke: sbuild-it just limits the # of params you need to pass, making it a simple sbuild-it *dsc | |
| [20:04] <blackboxsw> just helpful to cut down on the sbuild params. not really required or necessary | |
| [20:05] <blackboxsw> maybe unnecessary 'sugar' as this could nearly be solved with a simple bash alias | |
| [20:39] <blackboxsw> momousta: in landing your branch can I ammend your squashed commit message to the following: https://pastebin.ubuntu.com/p/7zRn9WpMqs/? | |
| [20:39] <blackboxsw> wanted to note that 410 is now being handle | |
| [20:39] <blackboxsw> wanted to note that 410 is now being handled | |
| [20:39] <momousta> Sure, please go ahead. | |
| [20:39] <blackboxsw> excellent, and done | |
| [20:39] <blackboxsw> thanks again | |
| [20:40] <momousta> Thanks | |
| [20:40] <blackboxsw> it'll be in our next upload to ubuntu groovy | |
| [21:18] <Odd_Bloke> blackboxsw: +1 on eoan | |
| [21:38] <Odd_Bloke> falcojr: So your PR is ready to merge, but I'm done for the day and don't have the time to write a good merge commit message (which I think is particularly important for this change). Would you mind putting one together for me to copy/paste first thing tomorrow (or perhaps by blackboxsw later)? | |
| [21:38] <falcojr> Odd_Bloke: Sure...I can do the squashing too | |
| [21:56] <blackboxsw> thanks Odd_Bloke uploaded that. and thanks lucasmoura uploaded bionic | |
| [21:57] <blackboxsw> just xenial remains for SRU upload, then I can ping the vanguard in #ubutu-devel to accept the queued uploads | |
| [22:01] <blackboxsw> xenial is here https://github.com/canonical/cloud-init/pull/406 | |
| [22:01] <blackboxsw> and I just uploaded 20.2-45 B, E and F to https://launchpad.net/~cloud-init-dev/+archive/ubuntu/proposed | |
| [22:01] <blackboxsw> so we should be able to test those bits directly in SRU validation | |
| [22:16] <lucasmoura> blackboxsw, I will review the xenial release right now | |
| [22:39] <lucasmoura> blackboxsw, we have a flake8 issue related to one of the existing patches in xenial | |
| [22:39] <lucasmoura> I have commented the issue on the PR | |
| [22:39] <blackboxsw> ugh thanks lucasmoura for both reviews | |
| [22:58] <blackboxsw> lucasmoura: for tomorrow I fixed and pushed the flake error | |
| [22:58] <blackboxsw> https://github.com/canonical/cloud-init/pull/406 | |