| lubot | [telegram] <teward001> right, anything else infra-wise that's busted and needs my attention | 00:44 |
|---|---|---|
| lubot | [telegram] <teward001> just checking in of course :) | 00:45 |
| * tsimonq2 sets up Chaos Monkey to keep teward busy | 01:28 | |
| arraybolt3[m] | OK, back in action. Simon Quigley I don't think I have the necessary access in Phab to do everything documented here: https://github.com/lubuntu-team/new-release | 16:06 |
| arraybolt3[m] | oh wait maybe I do? | 16:06 |
| arraybolt3[m] | Heh, nevermind, I do. Add that to the list of things I didn't know I was able to do. | 16:07 |
| arraybolt3[m] | And the stuff I probably don't have access to there I have access to in Gitea. | 16:11 |
| arraybolt3[m] | Testing wiki update complete. | 16:18 |
| arraybolt3[m] | (I may be slow since I'm juggling, but the process is in progress.) | 16:19 |
| arraybolt3[m] | Do we still want to be using the Phab sidebar, or should I be making that universal Gitea repo and starting to do stuff there instead? Simon Quigley, teward, Dan Simmons | 16:23 |
| arraybolt3[m] | OK so somehow I lost all of the branches of lubuntu-meta except for ubuntu/jammy... | 16:25 |
| * arraybolt3[m] digs | 16:26 | |
| arraybolt3[m] | Yep, just plain gone. Sigh. And I used git push --all... | 16:26 |
| arraybolt3[m] | Wow. So this is what I missed: https://stackoverflow.com/questions/10312521/how-do-i-fetch-all-git-branches And that is... insanely complicated. | 16:29 |
| arraybolt3[m] | At least it looks like it. | 16:29 |
| teward | fyi `git pull --all` exists | 16:32 |
| arraybolt3[m] | That doesn't do it. | 16:32 |
| arraybolt3[m] | My problem is that "git branch -r" shows me all the branches, and "git branch" shows me just one. | 16:32 |
| arraybolt3[m] | I need to pull every branch. | 16:33 |
| arraybolt3[m] | And now I just wiped my master command to do that... | 16:33 |
| tsimonq2 | git fetch --all | 16:36 |
| tsimonq2 | git pull --all likely replaces your local copies, git fetch is just updating what is considered to be the latest remote reference | 16:37 |
| * arraybolt3[m] sent a code block: https://libera.ems.host/_matrix/media/v3/download/libera.chat/c08e1d7ffb03bcdc51786e4d10e65aa2ffe55318 | 16:38 | |
| arraybolt3[m] | Bash makes my brain hurt. This is what finally did the trick: `for x in `git branch -r`; do git checkout -b $(printf "%s%s%s" $(echo $x | cut -d '/' -f2) "/" $(echo $x | cut -d '/' -f3)) $x; done` | 16:40 |
| arraybolt3[m] | Er... | 16:41 |
| arraybolt3[m] | for x in `git branch -r`; do git checkout -b $(printf "%s%s%s" $(echo $x | cut -d '/' -f2) "/" $(echo $x | cut -d '/' -f3)) $x; done | 16:41 |
| arraybolt3[m] | * Bash makes my brain hurt. This is what finally did the trick: mangled command here | 16:41 |
| arraybolt3[m] | Yes, that is two nested subshells within a subshell used in a for loop. | 16:41 |
| arraybolt3[m] | And yes, it throws some errors since the output of git branch -r isn't quite right but it's close enough for it to work. | 16:42 |
| tsimonq2 | If you're concerned about losing branches, just remember that we should still have at least a get up here and maybe even a launch pad mirror, if not fabricator itself even though all those services have been turned off | 16:43 |
| arraybolt3[m] | Talk to texans? | 16:44 |
| tsimonq2 | As a general rule I would tell you when you're pushing to a good repository that other people have to use, never use a command like get pulled all because it will have some negative unintended consequences | 16:44 |
| tsimonq2 | This was a fairly good case of that where it isn't that impactful | 16:44 |
| arraybolt3[m] | tsimonq2: OK. | 16:44 |
| tsimonq2 | To clarify now that I'm stopped... I'm glad this happened while we still have mirrors :P | 16:46 |
| arraybolt3[m] | I can't believe Git made it this hard. Like, really, is getting a full list of all branches that unusual? | 16:46 |
| arraybolt3[m] | Anyway this should be fairly easy to fix. | 16:47 |
| tsimonq2 | arraybolt3[m]: git fetch | 16:51 |
| tsimonq2 | fetch the status of the remote without pulling it (which clubbers your existing stuff) | 16:51 |
| tsimonq2 | s/clubbers/clobbers/ | 16:52 |
| arraybolt3[m] | Simon Quigley:... (full message at <https://libera.ems.host/_matrix/media/v3/download/libera.chat/013148df4a607924c26f42fb9618d099ef0c805a>) | 16:53 |
| tsimonq2 | git remote add mirror URL && git fetch --all && git push origin mirror/ubuntu/kinetic:ubuntu/kinetic (or similar) | 16:53 |
| tsimonq2 | one repo one clone less network needed :) | 16:54 |
| * tsimonq2 more coffee | 16:54 | |
| arraybolt3[m] | OH I GET IT. | 17:04 |
| arraybolt3[m] | Only just now did the last "git push" command click. | 17:04 |
| tsimonq2 | I would rather take this slow enough for you to learn than rush and do this for you - now you know how to deal with such a situation, I'm glad :) | 17:07 |
| tsimonq2 | Thanks for having enough patience with Git. I have a feeling you're past the Dunning-Kruger peak :P | 17:08 |
| arraybolt3[m] | LOL don't get your hopes up quite yet... I want to get there but this is really tricky. | 17:09 |
| arraybolt3[m] | Simon Quigley: OK so I'm still a bit confused here. | 17:09 |
| arraybolt3[m] | Simon Quigley: On Gitea, the only branch is ubuntu/jammy. | 17:10 |
| * tsimonq2 uploaded an image: (47KiB) < https://libera.ems.host/_matrix/media/v3/download/linuxdelta.com/QuzWEQOogLbqSiQzhpHolgYi/1024px-Dunning%E2%80%93Kruger_Effect_01.svg.png > | 17:10 | |
| arraybolt3[m] | On Phab, there's lots of branches. | 17:10 |
| tsimonq2 | arraybolt3[m]: The only branch or the default branch? | 17:10 |
| arraybolt3[m] | tsimonq2: The only branch. All the older ones are missing. | 17:10 |
| arraybolt3[m] | Also it looks like we neglected to make an ubuntu/kinetic branch last cycle, somehow. | 17:10 |
| tsimonq2 | arraybolt3[m]: Doesn't surprise me. It's okay, processes get defined this cycle :) | 17:10 |
| arraybolt3[m] | So I'm trying to get all of the older branches to migrate also. | 17:11 |
| arraybolt3[m] | I just want to mass-move all the branches from Phab onto Gitea. | 17:11 |
| lubot | [telegram] <Leokolb> Humm..now measuring 😊 metacognitive abilites... | 17:11 |
| tsimonq2 | here go https://stackoverflow.com/questions/37884832/git-push-all-branches-from-one-remote-to-another-remote | 17:11 |
| arraybolt3[m] | Ah, that looks more right. | 17:12 |
| arraybolt3[m] | (What I was trying to do is clone from Phab, change the origin URL to Gitea and then push. Now I'm realizing that's the wrong way to do that.) | 17:12 |
| tsimonq2 | @Leokolb: It's always in my head when training new people... "where are you on the curve?" I've seen enough of this to be patient :P | 17:12 |
| tsimonq2 | arraybolt3[m]: It's not the wrong way to do it necessarily, it's just easier and more "best practice" to do it this way instead | 17:13 |
| tsimonq2 | (Add the remote instead of replacing) | 17:13 |
| arraybolt3[m] | Perfect. Alright, let's try it. | 17:13 |
| * arraybolt3[m] starts playing MercyMe's "We Win" | 17:15 | |
| arraybolt3[m] | OK so now I guess I do that to the other three migrated repos. | 17:16 |
| arraybolt3[m] | Turns out lubuntu-meta was the only repo with multiple branches. | 17:23 |
| arraybolt3[m] | So I guess that's taken care of now. | 17:23 |
| arraybolt3[m] | (The only repo that was migrated the other day, I mean.) | 17:23 |
| tsimonq2 | Very nice! Thank you :) | 17:33 |
| tsimonq2 | TIL `set -eux && echo $DEBEMAIL` crashes QTerminal :P | 17:34 |
| arraybolt3[m] | OK I just had yet more personal junk hit, but not as severe as last time. Probably going to be gone for about an hour or so, brb asap. | 17:37 |
| arraybolt3[m] | Alright that took forever. | 19:06 |
| arraybolt3[m] | lubuntu-meta update.cfg reconfigured for Lunar. | 19:14 |
| tsimonq2 | arraybolt3[m]: If it builds, let me know, I'll upload | 19:14 |
| tsimonq2 | (I'm willing to bet though, unless you grab my debootstrap from the Lunar queue, download, build, and install it, that it will fail) | 19:15 |
| arraybolt3[m] | Oh hold on it's not that far done. | 19:15 |
| arraybolt3[m] | (I checked gitk to see how it was done and missed badly now that I'm looking at the changelog...) | 19:15 |
| tsimonq2 | No worries, take as much time as you need | 19:15 |
| arraybolt3[m] | OK I'm starting to get it. (debootstrap from the Lunar queue?) | 19:16 |
| tsimonq2 | https://launchpad.net/ubuntu/lunar/+queue | 19:17 |
| tsimonq2 | `Show` -> `Unapproved` | 19:17 |
| tsimonq2 | `Search` | 19:17 |
| tsimonq2 | * `Update` | 19:17 |
| arraybolt3[m] | Yikes, wait, I'm supposed to install that on the system that does the building? | 19:17 |
| arraybolt3[m] | If so, /me needs time to spin up a VM | 19:18 |
| tsimonq2 | well I'm the one that authored the upload, let me tell you how to jerry-rig it for your system ;) | 19:18 |
| tsimonq2 | symlink `/usr/share/debootstrap/scripts/gutsy` -> `/usr/share/debootstrap/scripts/lunar` | 19:18 |
| tsimonq2 | and debootstrap will magically work | 19:18 |
| arraybolt3[m] | And that's after installing Lunar's debootstrap on my system, or I just do that with the one I have here? | 19:19 |
| tsimonq2 | You can just do that with the one you have installed | 19:19 |
| tsimonq2 | (That's all the upload does; create that symlink) | 19:19 |
| arraybolt3[m] | Oh! I get it! So eventually that upload will end up in all Ubuntu systems anyway, right? | 19:20 |
| arraybolt3[m] | (I am going to make the symlink, I'm just making sure I understand how things work here.) | 19:20 |
| tsimonq2 | arraybolt3[m]: Right :) | 19:22 |
| tsimonq2 | (Although, I don't often see debootstrap SRUs) | 19:22 |
| arraybolt3[m] | 👍️, ok, cleaning up loose ends and then here goes! | 19:23 |
| arraybolt3[m] | Simon Quigley: Just to be clear on something, I see that for each release except the ones that got skipped, there's a "Welcome to Hirsute", "Welcome to Jammy', etc. commit in Git. I assume that commit isn't made until the release is put out, right? | 19:24 |
| arraybolt3[m] | Like I'm not about to make a "Welcome to Lunar" commit. | 19:24 |
| tsimonq2 | Your assumption is incorrect :) | 19:27 |
| tsimonq2 | Many of the uploads you've just seen from me have the text "Welcome to the Lunar Lobster!" | 19:27 |
| tsimonq2 | "Welcome" messages are usually done at the beginning of the cycle, is what I'm trying to say | 19:27 |
| arraybolt3[m] | Ah. That's why I asked. :) | 19:28 |
| arraybolt3[m] | I just noticed lots of changes that lookedl ike they were made late in the cycle (version number changes and the like) in the "Welcome to jammy" commit. | 19:28 |
| arraybolt3[m] | So... how exactly does that work? Are those changes made early in the cycle? | 19:28 |
| tsimonq2 | I can only speak for my time as Release Manager and my governance of such things - I personally take the approach of "as soon as we release the latest one, jump on as much as you possibly can, get it all updated, and we'll be able to spend more time working out the kinks and less time updating version numbers" | 19:29 |
| tsimonq2 | That's also part of the reason I decided to personally get involved in the Qt 5.15.7 transition; the more time we have to find Qt bugs, the more mature things tend to be when it actually is crunch time :) | 19:30 |
| tsimonq2 | That's my entire point of "hey, yo, everyone take the weekend off" because someone needs to hold the executive position, I'm not comfortable putting some of this work on anyone else | 19:31 |
| tsimonq2 | It's all documented, so you should be able to follow in my steps, but it's a huge burden | 19:31 |
| arraybolt3[m] | Makes sense. But... I don't know what we want Lunar's seed to look like this cycle. | 19:31 |
| arraybolt3[m] | I mean, in the "Welcome to Jammy" commit, we removed Trojita (I think) and stuff like that. So I guess the question is, what all should I add and remove right now? | 19:32 |
| arraybolt3[m] | Either that or else I'm badly confused and still need a bit of extra explanation. | 19:32 |
| tsimonq2 | All of the lists in lubuntu-meta are automatically generated from the contents of the seed (see historical context, I made sure, a few years ago, that this was all Git and the main Ubuntu infrastructure supported it as such): https://lubuntu.me/lubuntu-seeds-are-now-in-git/ | 19:33 |
| tsimonq2 | lubuntu-meta should be updated every time the seed is updated, both the common Ubuntu seed and Lubuntu's | 19:34 |
| tsimonq2 | Is that done in practice? lol nah | 19:34 |
| tsimonq2 | That's where your confusion may come from | 19:34 |
| tsimonq2 | "Welcome to Jammy" was just likely the first update they've done in a while | 19:34 |
| tsimonq2 | * "Welcome to Jammy" was just likely the first update they did that cycle | 19:35 |
| arraybolt3[m] | click I get it! | 19:35 |
| arraybolt3[m] | So I'm supposed to run whatever automatically updates those lists and then document the changes it makes in the changelog? | 19:35 |
| tsimonq2 | I'd say yes, but I don't want you to get confused: the script even updates the changelog for you :) | 19:35 |
| arraybolt3[m] | Oh. :P Alright so it really is just "update that one file and then trigger everything". | 19:36 |
| tsimonq2 | yep :) | 19:36 |
| arraybolt3[m] | So do I now run `./update`? | 19:36 |
| arraybolt3[m] | (I'm in the lubuntu-meta Git repo.) | 19:37 |
| tsimonq2 | *Usually*, yes. First, I'd make sure the Git repo is in sync with the archive and `git log` is full of nice tags | 19:38 |
| arraybolt3[m] | Oh sap. That probably isn't true. | 19:38 |
| tsimonq2 | (Make sure your baseline is established and in-sync with the archive before you make the changes, is what I'm saying.) | 19:38 |
| arraybolt3[m] | Right. I always forget to do that. | 19:38 |
| tsimonq2 | Well, you can create tags, that's no big deal, we've never required signed tags (that's a Fedora thing) | 19:38 |
| lubot | [telegram] <kc2bez> Sometimes late people are late :/ (re @lubuntu_bot: (irc) <tsimonq2> * "Welcome to Jammy" was just likely the first update they did that cycle) | 19:40 |
| tsimonq2 | That's why I'm not taking the time to rail on you guys for it, it happens, you know? I understand, not having me or Aaron, you guys were two people less, if that makes sense | 19:41 |
| tsimonq2 | At least it got done eventually :) | 19:41 |
| tsimonq2 | And I will continue to express my appreciation for you guys holding down the fort during my hiatus | 19:42 |
| lubot | [telegram] <kc2bez> Yeah, not necessarily a proud moment for me. | 19:42 |
| tsimonq2 | Such is life. No shame from me :) | 19:42 |
| arraybolt3[m] | OK so shoot. How do I delete just a branch on Gitea? | 19:42 |
| arraybolt3[m] | (I made the ubuntu/lunar branch a bit too early.) | 19:42 |
| tsimonq2 | arraybolt3[m]: `git push REMOTE_NAME :MY_BRANCH` | 19:43 |
| tsimonq2 | so in your case, *probably*, `git push origin :ubuntu/lunar` | 19:43 |
| tsimonq2 | (`null` before `:` means delete the remote branch you're about to specify) | 19:43 |
| arraybolt3[m] | Nice, thanks! | 19:43 |
| tsimonq2 | Of course :) | 19:43 |
| * tsimonq2 looks like a homeless person right now - about to go get my hair and beard cut :) | 19:45 | |
| * tsimonq2 afk for probably an hour | 19:45 | |
| arraybolt3[m] | o/ | 19:45 |
| arraybolt3[m] | Ah, now not having an SSH key on my Launchpad profile has come back to bite me. | 19:50 |
| arraybolt3[m] | s/Ah/ArGH!/ | 19:50 |
| arraybolt3[m] | Hey it's finally working! \o/ | 20:14 |
| arraybolt3[m] | Sooo... I don't have a Lunar schroot that I can use to build this. | 20:19 |
| arraybolt3[m] | Do I just build it with Kinetic and hope for the best? | 20:20 |
| arraybolt3[m] | FWIW it builds with Kinetic just fine. | 20:21 |
| arraybolt3[m] | Also Lintian makes it look like I might should update some of the packaging, is that correct? | 20:29 |
| arraybolt3[m] | It's been two and a half years since we made any packaging changes to lubuntu-meta other than messing with the changelog. I'm going to declare it a good time to upgrade our packaging there if no one has any objections. | 21:43 |
| arraybolt3[m] | (Last it was updated was before Focal, looks like. It may have been longer than two and a half years.) | 21:43 |
| tsimonq2 | <arraybolt3[m]> "Do I just build it with Kinetic..." <- Pragmatically speaking: the underlying toolchain and packages have not changed at all during this point of the cycle. FWIW, I was able to create a Lunar schroot using the following syntax, after the local symlink dance: `mk-sbuild lunar --distro="ubuntu"` | 21:48 |
| tsimonq2 | "Yes, you can technically still use a Kinetic schroot at this point in the cycle, but harumph, while it walks like a Kinetic it doesn't quack like a Kinetic, so it's not a Kinetic" | 21:49 |
| tsimonq2 | s/Kinetic/Lunar/, s/Kinetic/Lunar/, s/Kinetic/Lunar/ | 21:49 |
| tsimonq2 | <arraybolt3[m]> "It's been two and a half years..." <- Exactly the best time :) | 21:50 |
| arraybolt3[m] | Nice. OK, looks like we're all set then. | 21:50 |
| arraybolt3[m] | (We're still on Debian Policy Manual version 4.1.5...) | 21:50 |
| tsimonq2 | <arraybolt3[m]> "(We're still on Debian Policy..." <- Best practice: read all the changelogs from Debian Policy and check all of them | 22:21 |
| arraybolt3[m] | Simon Quigley: I just used this extremely handy tool: https://www.debian.org/doc/debian-policy/upgrading-checklist.html#version-4-1-5 | 22:21 |
| tsimonq2 | Precisely, nice ;) | 22:22 |
| arraybolt3[m] | Didn't end up needing a single change, miraculously. | 22:22 |
| arraybolt3[m] | (And I even dug deeper into a couple of spots where I wasn't quite sure.) | 22:22 |
| arraybolt3[m] | Whew! I think I'm done. | 22:24 |
| arraybolt3[m] | Just rebuilding to make sure everything worked. | 22:25 |
| arraybolt3[m] | Simon Quigley: One small problem, lubuntu-desktop depends on an X fonts package, which Debian's manual says must not happen. It will let me put it as a Recommends, though. Trouble is it must be sneaking in through ${misc:Depends}. Is there some way for me to specifically exclude a Depends, or is this an indication of something broken elsewhere? | 22:29 |
| arraybolt3[m] | (The particular package is `xfonts-efont-unicode`) | 22:30 |
| tsimonq2 | Can I please get a link to the policy stanza you're referring to so I can figure out intent? | 23:03 |
| arraybolt3[m] | Simon Quigley: https://www.debian.org/doc/debian-policy/ch-customized-programs.html, section 11.8.5. | 23:04 |
| arraybolt3[m] | Point 1. | 23:04 |
| arraybolt3[m] | The link to Footnote 8 contains the reasoning. | 23:04 |
| tsimonq2 | Yeah... I think you're right on this one | 23:06 |
| tsimonq2 | They are very intentional with must/should wording, I Shall Not Doubt Debian Policy | 23:07 |
| tsimonq2 | Please create an MP on Launchpad for the seed/throw me a Git patch and I'll be happy to get it in... all you have to do is surround that font package in parentheses | 23:07 |
| tsimonq2 | VERY NICE catch | 23:08 |
| arraybolt3[m] | I don't know what an MP means but sure. | 23:08 |
| tsimonq2 | Merge Proposal :) | 23:08 |
| tsimonq2 | (Same as a PR) | 23:08 |
| tsimonq2 | Also, changes here would be done in the seed, not the control file itself :) | 23:08 |
| arraybolt3[m] | You read my mind, that's what I was just about to ask. | 23:08 |
| tsimonq2 | Sweet :) | 23:08 |
| tsimonq2 | I need to get going here, but I'll check in later | 23:09 |
| arraybolt3[m] | <tsimonq2> "Please create an MP on Launchpad..." <- ``... (full message at <https://libera.ems.host/_matrix/media/v3/download/libera.chat/5fd2d20cfbfa756cfaea4458a4d71fed323ec380>) | 23:16 |
| arraybolt3[m] | > <@tsimonq2:linuxdelta.com> Please create an MP on Launchpad for the seed/throw me a Git patch and I'll be happy to get it in... all you have to do is surround that font package in parentheses... (full message at <https://libera.ems.host/_matrix/media/v3/download/libera.chat/df8aa974dee338d8578aeb4e3dcecfdb0892283a>) | 23:16 |
| arraybolt3[m] | * ```... (full message at <https://libera.ems.host/_matrix/media/v3/download/libera.chat/ef55c9d566b5860756a3f97cf1f362ec65071c40>) | 23:17 |
| arraybolt3[m] | * ```... (full message at <https://libera.ems.host/_matrix/media/v3/download/libera.chat/b3c530f2d15cff55f64eef3b8922de8b2899ad35>) | 23:17 |
| tsimonq2 | That's not verbose :/ I'll look in a bit heh | 23:32 |
| arraybolt3[m] | OK. I'll send you what I have so far shortly. | 23:32 |
| * arraybolt3[m] posted a file: (9KiB) < https://libera.ems.host/_matrix/media/v3/download/matrix.org/fOwfmUkUQYaxYDZQGFlQXvro/diff.txt > | 23:33 | |
| arraybolt3[m] | Simon Quigley: ^ diff containing all my changes to lubuntu-meta, THIS DOES NOT BUILD | 23:34 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
