| TJ- | Right! initramfs-tools had a MAJOR import from Debian with "initramfs-tools (0.133ubuntu1)" April 29th; I've worked back through the changes by Debian to "initramfs-tools (0.132) unstable" of 26th July 2018, which contains "[f39625a] update-initramfs: Make "-k all" take over other initramfs images" and looking at that commit seems to indicate it is the cause | 00:02 |
|---|---|---|
| TJ- | https://salsa.debian.org/kernel-team/initramfs-tools/commit/f39625afd6ba6c1aa2027286dc3ef1c933da14e0 | 00:02 |
| TJ- | I can reproduce the failure calling the command with the same options | 00:03 |
| wxl | huh | 00:04 |
| lubot | <kc2bez> I was just cruising the changelog on that. I hadn't got to that yet. There were a ton of changes that got merged. | 00:04 |
| TJ- | adding the --verbose flag shows: | 00:05 |
| TJ- | update-initramfs -k all -c -t -v | 00:05 |
| TJ- | Available versions: | 00:05 |
| TJ- | Nothing to do, exiting. | 00:05 |
| TJ- | here's the git blame of the file, scroll to get_sorted_versions() https://salsa.debian.org/kernel-team/initramfs-tools/blame/master/update-initramfs | 00:06 |
| wxl | is it calamares that introduces -k all where ubiquity does not? | 00:07 |
| wxl | because i'm also trying to wrap my head around why this is only us | 00:07 |
| TJ- | wxl: not as far as I can tell; that has been in the calamares initramfs module for a long time | 00:07 |
| TJ- | wxl: the problem here is the code is *expecting* there to be existing initrd.img files when there won't be in this situation. That would make sense for the -u --update option but not for -c --create | 00:08 |
| wxl | i see what you mean | 00:09 |
| TJ- | I turned shell recording on, you can see what it does here: https://paste.ubuntu.com/p/V8C7Js5RBY/ | 00:09 |
| TJ- | what a rabbot-hole this has turned out to be! | 00:10 |
| wxl | no kidding | 00:10 |
| wxl | but i'm so glad to have you on this one, TJ-. i'm not as much of a low level wizard as you. i'd probably be able to get myself there but it would take about 10x the effort | 00:11 |
| TJ- | wxl: I have a kludge workaround we can deploy via calamares config until initramfs-tools is fixed upstream | 00:12 |
| wxl | TJ-: so why doesn't this affect ubiquity again?? | 00:12 |
| TJ- | it consists of "touch /boot/initrd.img-5.0.0-17-generic" (just need it to get the kernel version dyanmically of course) | 00:13 |
| wxl | oh what a kludge it is XD | 00:13 |
| TJ- | wxl: because presumably it does things differently ( doesn't use "-k all" I'd guess) | 00:13 |
| wxl | right, that's what i was guessing | 00:13 |
| TJ- | I've proved that works, now just need to add it to calamares | 00:13 |
| TJ- | finally! | 00:14 |
| wxl | i guess this should be in the default cala config.. i know kubuntu's been talking about using it | 00:14 |
| TJ- | Building cpio /boot/initrd.img-5.0.0-17-generic.new initramfs | 00:15 |
| wxl | if you could leave a comment in there referencing the bug, that would be great. memories are only so long :) | 00:15 |
| TJ- | I am using the bug number as the name of the module instance :) | 00:17 |
| wxl | heheheh | 00:17 |
| * TJ- fingers crossed | 00:20 | |
| TJ- | Yes! | 00:28 |
| TJ- | root@lubuntu:/etc/calamares# ls -l /target/boot/initrd.img* | 00:28 |
| TJ- | lrwxrwxrwx 1 root root 27 Jun 18 17:41 /target/boot/initrd.img -> initrd.img-5.0.0-17-generic | 00:28 |
| TJ- | -rw-r--r-- 1 root root 0 Jun 19 01:22 /target/boot/initrd.img-5.0.0-17-generic | 00:28 |
| TJ- | lrwxrwxrwx 1 root root 27 Jun 18 17:41 /target/boot/initrd.img.old -> initrd.img-5.0.0-17-generic | 00:28 |
| wxl | you are my hero TJ- | 00:28 |
| kc2bez | Mine too | 00:29 |
| TJ- | There are a load of errors when trying to reboot with it failing to unmount /cdrom - are those being addressed? | 00:30 |
| TJ- | looks like the squashfs gets ripped away from under it and it gets messed up | 00:30 |
| TJ- | See https://iam.tj/projects/ubuntu/lubuntu-installer-restart.png | 00:32 |
| wxl | oh interesting | 00:32 |
| wxl | hadn't seen that before | 00:32 |
| TJ- | I've had it happen a few times during testing; have to do a forced reset of the VM | 00:33 |
| TJ- | it could be related to my working in the shell but I can't think how it'd cause that | 00:33 |
| kc2bez | Oh geez, I haven't encountered that either. | 00:33 |
| TJ- | Hmmm, the VM seems to have rebooted but there's a blank screen, no serial terminals, and CPU pegging 100% | 00:35 |
| TJ- | well now, this is weird, it's hitting the same panic! | 00:37 |
| TJ- | I'm far too tired! None of us spotted what I pasted above! | 00:38 |
| TJ- | -rw-r--r-- 1 root root 0 Jun 19 01:22 /target/boot/initrd.img-5.0.0-17-generic | 00:38 |
| wxl | oh no | 00:38 |
| * wxl cries | 00:39 | |
| TJ- | !! the touched file | 00:39 |
| ubot93 | Factoid '! the touched file' not found | 00:39 |
| TJ- | no size | 00:39 |
| TJ- | so it still failed :( | 00:39 |
| wxl | well thanks for the effort | 00:39 |
| TJ- | I've not finished ! | 00:39 |
| wxl | dude get some sleep though :) | 00:39 |
| * TJ- never lets the bug win :p | 00:41 | |
| wxl | well i won't complain :) | 00:43 |
| lubot | <HMollerCl> @tsimonq2 wxl running update script on meta give this: | 00:53 |
| lubot | <HMollerCl> ./update: 13: exec: germinate-update-metapackage: not found | 00:53 |
| lubot | <HMollerCl> I only need to clone lubuntu-meta? or something else is also needed? | 00:53 |
| lubot | <UniversalSuperBox> https://manpages.debian.org/stretch/germinate/germinate-update-metapackage.1.en.html | 00:53 |
| lubot | <UniversalSuperBox> ;) | 00:53 |
| lubot | <HMollerCl> thanks! | 01:02 |
| TJ- | This gets weirder... the reason update-initramfs get_sorted_versions() fails is its call to (/usr/bin/) "linux-version list" reports nothing. That's a Perl script that eventually calls /usr/share/perl5/DebianLinux.pm::image_list() to scan /boot/ for kernel image (/boot/vmlinuz-*) | 01:22 |
| lubot | <HMollerCl> wxl @tsimonq2 afetr running update script (and installing germinate) https://paste.ubuntu.com/p/trjJZTrfS8/ | 01:31 |
| lubot | <tsimonq2> wat? | 01:31 |
| lubot | <tsimonq2> Hmm | 01:31 |
| lubot | <HMollerCl> (this change in chromium from apt to snap has been very painfull) | 01:33 |
| TJ- | OK, this time I've really figured it out :D | 01:34 |
| TJ- | I blame whoever wrote lubuntu/settings.conf :p | 01:34 |
| lubot | <tsimonq2> Oh? | 01:36 |
| lubot | <tsimonq2> @HMollerCl [(this change in chromium from apt to snap has been very painfull)], For sure | 01:36 |
| TJ- | How do I add a patch/diff via phabricator? I'm getting lost in the UI | 01:40 |
| lubot | <tsimonq2> It's a CLI client | 01:40 |
| lubot | <tsimonq2> Off the top of my head: … 1. `git clone REPO` … 2. Make the changes in the repo. Add the files but don't commit. … 3. `arc diff` (follow the prompts) | 01:41 |
| wxl | arcanist | 01:41 |
| lubot | <tsimonq2> Make sure the Reviewers section is `@wxl @tsimonq2` | 01:41 |
| TJ- | tsimonq2: so I need to install 'arc' ? | 01:42 |
| lubot | <HMollerCl> @tsimonq2 [For sure], look at this: | 01:42 |
| wxl | yep | 01:42 |
| wxl | ^ TJ- | 01:42 |
| lubot | <HMollerCl> sudo apt autoremove … Reading package lists... Done … Building dependency tree … Reading state information... Done … The following additional packages will be installed: … chromium-browser … The following packages will be REMOVED: … chromium-browser-l10n chromium-codecs-ffmpeg-extra libpoppler85 linux-headers-5.0.0-13 lin | 01:42 |
| lubot | ux-headers-5.0.0-13-generic … linux-headers-5.0.0-8 linux-headers-5.0.0-8-generic linux-image-5.0.0-13-generic linux-image-5.0.0-8-generic … linux-modules-5.0.0-13-generic linux-modules-5.0.0-8-generic linux-modules-extra-5.0.0-13-generic … linux-modules-extra-5.0.0-8-generic … The following packages will be upgraded: … chromium-browser … 1 | 01:42 |
| lubot | upgraded, 0 newly installed, 13 to remove and 0 not upgraded. … 2 not fully installed or removed. … Need to get 0 B/48,4 kB of archives. … After this operation, 889 MB disk space will be freed. … Do you want to continue? [Y/n] | 01:42 |
| lubot | <HMollerCl> sorry | 01:42 |
| wxl | TJ-: btw you can blame tsimonq2 mostly XD https://phab.lubuntu.me/source/calamares-settings-ubuntu/browse/master/lubuntu/settings.conf | 01:42 |
| lubot | <HMollerCl> the thing is, chromium wants to install it snap and I don't want it, but it won't let me do anything until I install snap | 01:43 |
| lubot | <tsimonq2> @HMollerCl [the thing is, chromium wants to install it snap and I don't want it, but it won' …], Grab the package from Debian | 01:44 |
| wxl | or ditch chromium XD | 01:44 |
| * wxl ducks | 01:44 | |
| lubot | <HMollerCl> @tsimonq2 [Grab the package from Debian], will try that | 01:44 |
| wxl | ah ha! | 01:45 |
| wxl | @tsimonq2 screwed up again | 01:45 |
| wxl | problem is right here: https://phab.lubuntu.me/source/lubuntu-meta/browse/ubuntu%252Feoan/update.cfg$4 | 01:45 |
| wxl | and here https://phab.lubuntu.me/source/lubuntu-meta/browse/ubuntu%252Feoan/update.cfg$16 | 01:46 |
| wxl | should probably remove i386 while you're at it | 01:46 |
| wxl | headed home talk soon | 01:47 |
| lubot | <teward001> @wxl [<wxl> @teward001 do you know where "white_listed_spam_host_domains" is defined o …], I believe it has to be defined in the configuration file somewhere, but i'll have to check | 01:47 |
| lubot | <HMollerCl> I installed a .deb in chromium now is worst | 02:04 |
| TJ- | ooo looking hopeful, its hanging about saying "Creating initramfs" this time | 02:06 |
| TJ- | and its there, finally! | 02:07 |
| TJ- | problems, arcanist complaining php-curl needs installing, but it is installed¬ | 02:31 |
| lubot | <HMollerCl> sigh! "chromium-browser … package is in a very bad inconsistent state; you should reinstall it before attempting a removal" | 02:32 |
| lubot | <tsimonq2> @TJ- [<TJ-> problems, arcanist complaining php-curl needs installing, but it is instal …], :/ hm | 02:39 |
| wxl[m] | TJ- php-cli installed? | 02:41 |
| TJ- | wxl[m]: yes, it's a depends of arcanist (and a Provides: of php7.2-cli) | 02:44 |
| wxl[m] | TJ-: fwiw I always use upstream arcanist | 02:47 |
| wxl[m] | https://secure.phabricator.com/book/phabricator/article/arcanist_quick_start/ | 02:48 |
| wxl[m] | In other words I use git master | 02:48 |
| TJ- | OK, solved that! It installed the phph7.2-curl but php7.0-cli! added php7.2-cli and it runs, but now I have to give it a phabricator URI!? | 02:50 |
| lubot | <tsimonq2> https://phab.lubuntu.me/ | 02:51 |
| TJ- | Is this something I need to add into the commit message? | 03:04 |
| TJ- | Make sure the Reviewers section is `@wxl @tsimonq2` | 03:04 |
| wxl[m] | No there will be a template | 03:04 |
| lubot | <tsimonq2> Two separate files to save | 03:05 |
| lubot | <tsimonq2> First is the commit message, second is the filled out template | 03:05 |
| TJ- | Ahhh, thanks! I didn't want to save the commit message and find out they were supposed to be in it! | 03:07 |
| TJ- | Wow, didn't take long did it (checks what time we started 20.09..04:10) https://phab.lubuntu.me/D15 | 03:10 |
| lubot | <tsimonq2> ULTIMATE bonus points for bug name | 03:11 |
| lubot | <tsimonq2> I like it! | 03:11 |
| lubot | <tsimonq2> Does it work? | 03:12 |
| TJ- | tsimonq2: I had to add a Test Plan statement! | 03:12 |
| lubot | <tsimonq2> Some people do that and still don't test it ;) | 03:12 |
| TJ- | Please have others test this before committing though | 03:12 |
| lubot | <tsimonq2> @kc2bez wxl ^ | 03:13 |
| TJ- | I've lost track of how many times I re-ran the VM install to get to the root of the issue | 03:13 |
| lubot | <tsimonq2> Thank you! | 03:14 |
| lubot | <tsimonq2> Your work is very very appreciated | 03:14 |
| TJ- | It's easy enough to manually edit the /etc/calamares/ files once the live env has started, before running the installer itself | 03:14 |
| lubot | <tsimonq2> Want to go the extra mile and look at fixing update-initramfs, or should I do that one? | 03:15 |
| TJ- | tsimonq2: That's going to have to be done in Debian I think, by those that know the code intimatedly. I cannot be sure its wrong but it certainly seems to be, when using -c (create) | 03:17 |
| TJ- | Ben Hutchins wrote that; I'm not going to argue with him! | 03:17 |
| lubot | <tsimonq2> This sounds like cjwatson/vorlon/infinity territory | 03:18 |
| lubot | <tsimonq2> It's like a bingo card. I'll bet one of them at least is a comaintainer | 03:18 |
| TJ- | the problem is its hard to know what the logic in that part of the code is supposed to do, it seems to only be relevant for an update operation but ...! | 03:19 |
| * TJ- is off to bed finally. Will drop in later today (Wednesday) to see if its solved the issue for everyone | 03:30 | |
| lubot | <tsimonq2> Thanks! | 03:34 |
| wxl | @kc2bez: it worked for me!!! please double check and i think we're back in action | 04:40 |
| * wxl pulls his hair out trying to find where the code is for usb-creator because none of the repos seem to include 0.3.6 | 06:20 | |
| kc2bez | wxl: @tsimonq2 It worked for me too! | 10:29 |
| kc2bez | Many many thanks to TJ- !! | 10:29 |
| wxl | anyone else want to test https://phab.lubuntu.me/D15? | 20:12 |
| TJ- | wxl g'evening | 20:39 |
| wxl | TJ-: afternoon | 20:40 |
| lubot | <SamuelBanya> Are there any non qt packages that need to be ported? | 21:26 |
| wxl | that would be a bit counterintuitive | 21:26 |
| lubot | <HMollerCl> @SamuelBanya [Are there any non qt packages that need to be ported?], The porting we need is from gtk to qt. I can't think of any other porting we need. | 21:28 |
| lubot | <HMollerCl> I would like to port some things from python to c++, but that's me and I think it would be harder than port gtk to qt. | 21:29 |
| lubot | <SamuelBanya> I only ask because it's still a bit intense to wrap my mind on not only gtk to qt but to try to use pyqt when the docs favor c++ | 21:29 |
| lubot | <SamuelBanya> I'm not the greatest at qt honestly. Never liked it in that respect and it's a bit hard to be honest | 21:30 |
| lubot | <SamuelBanya> Anyway I still haven't heard back from that ubuntu dev regarding the iso tester, will ping him again on LinkedIn to ask if he got my email | 21:31 |
| lubot | <lynorian> Do we still need testing for the whole initramfs stuff | 21:38 |
| wxl | if you could please i would appreciate it @lynorian | 21:38 |
| wxl | i just want to be sure | 21:38 |
| lubot | <lynorian> so what do I need to do to test this? | 21:38 |
| wxl | to be fair, i only tested bios/unencrypted on a vm. we should really test EVERYTHING, just to be extra sure | 21:38 |
| wxl | https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1829805/comments/17 | 21:39 |
| ubot93 | Ubottu bug 1829805 in initramfs-tools (Ubuntu) "Lubuntu Eoan Daily Image fails to boot after install on KVM" [Undecided, Confirmed] | 21:39 |
| TJ- | wxl: you want the other modes/configs tested (efi/LUKS) ? | 21:39 |
| wxl | TJ-: yes please. and real hardware is always good, too | 21:43 |
| TJ- | best to respin the squashfs.filesystem then :) | 21:44 |
| wxl | vm/bios/luks is a go | 23:14 |
| lubot | Qusai was added by: Qusai | 23:15 |
| wxl | vm/efi/unencrypted is a go | 23:31 |
| wxl | vm/efi/luks is a go | 23:44 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
