| Wizzup | freemangordon: btw, this will surface chimaera-era packages after dist upgrade: dpkg -l | grep '+m7' | 00:15 |
|---|---|---|
| Wizzup | https://phoenix.maemo.org/view/Images/job/leste-image-n900/232/console | 11:10 |
| Wizzup | n900 image built | 11:10 |
| Wizzup | I will disable tmpfs again now that I think I solved the I/O issue | 11:10 |
| Wizzup | well, it still took 8 hours :) | 11:10 |
| Wizzup | but that's better than nothing for now! | 11:10 |
| Wizzup | (above is not daedalus fyi) | 11:11 |
| arno11 | Wizzup: cool @chimaera img | 13:27 |
| arno11 | i see it is still kernel 6.1 which is good to avoid freeze on first boot :) | 13:27 |
| arno11 | is devel stuff almost included in the img (like daedalus) or ? | 13:28 |
| Wizzup | most of it should be, my next step is to look at your MRs and merge to daedalus stable | 13:29 |
| Wizzup | and I guess the main n900 image problem now for daedalus is the swap | 13:29 |
| Wizzup | image-builder overwriting the leste-config | 13:29 |
| arno11 | yes indeed | 13:30 |
| arno11 | and yes this is the main issue (swap) | 13:30 |
| Wizzup | yeah, this is a bit annoying since it's all /etc/fstab and postinst magic | 13:31 |
| Wizzup | there is no /etc/fstab.d or whatever | 13:31 |
| Wizzup | let me see what I can do afte | 13:31 |
| Wizzup | s/afte// | 13:31 |
| arno11 | ok | 13:31 |
| Wizzup | maybe we run the leste-config configure again at the end of something | 13:54 |
| Wizzup | that seems like a sensible generic solution imo | 13:55 |
| donihalim | df lh | 14:03 |
| arno11 | Wizzup: ok | 14:12 |
| kiva | yellow exclamation mark updates done today..worked in Pinephone. | 14:44 |
| kiva | although..still clock crash if I try change time. | 14:47 |
| Wizzup | is there a github issue/bug report for this? | 14:49 |
| Wizzup | otherwise I am not sure why some meta pkg reorderings would fix this | 14:51 |
| kiva | it is that old problem..and problem might be somewhere else..it is mentioned in wiki and there is how to set timezone. | 14:53 |
| Wizzup | please file a bug report with reproducible steps if there is no issue yet | 14:56 |
| kiva | I think it is known bug. Just start Clock app and just touch clock time and app crash. | 14:58 |
| Wizzup | works fine for me. | 14:59 |
| Wizzup | please file a bug report with the device, distro version, and steps to reproduce | 14:59 |
| arno11 | works fine for me as well | 15:00 |
| kiva | really? When next sd-card image for Pinephone comes, it would be best that I test with fresh image. | 15:01 |
| arno11 | yeah that's probably a good idea | 15:03 |
| kiva | wait a minute there is new image dated today :) | 15:14 |
| arno11 | Wizzup: i opened few issues for Daedalus (not critical, just as a reminder: osso_games, sdl, qt5 launch) | 15:49 |
| Wizzup | ty! | 15:55 |
| Wizzup | btw, any opinions on (eventually) moving away from github to our own hosted env? | 15:56 |
| Wizzup | I thought github would make contributing much easier, but I don't know that really is the case | 15:57 |
| uvos__ | i would say github helps discovery if nothing else | 16:01 |
| Wizzup | we can still keep the repos, just as read-only mirror or something | 16:05 |
| Wizzup | anyway just an idea, we have bigger fish to try any way | 16:05 |
| Wizzup | but if we do go for says gogs we'd have the ability to control our ci a lot more | 16:06 |
| inky | uvos__, what should i do to disable hope, search etc leds on bionic? in the night those are very bright, brighter than the screen, i cover them with finger to read. on d4 those are fine and not noticable. | 17:05 |
| inky | s/hope/home/ | 17:05 |
| sicelo | Wizzup: re - move away from GH - "I thought github would make contributing much easier, but I don't know that really is the case" - what issues have you observed? | 17:39 |
| Wizzup | I haven't seen a lot of people come to github before coming to irc :) | 17:39 |
| Wizzup | the thought was that having to make an account [to another service] would be a blocker to contributing | 17:40 |
| Wizzup | and also tbh, I've never liked github much, and I'd rather move away from the US-centric service | 17:40 |
| sicelo | alright. of course an own env is an extra thing to maintain | 17:41 |
| sicelo | postmarketOS has moved around quite a bit: github.com -> gitlab.com -> sr.ht -> gitlab.com -> gitlab.postmarketos.org (hosted by OSL) ... | 17:47 |
| arno11 | moving away from US-centric services makes sense for a foss project in 2025, imo | 17:51 |
| sicelo | i am not against a move, but not sure it's *that* important | 17:54 |
| sicelo | arno11: while linux kernel remains under US legislation :-p (hence a couple of months ago they stripped Russian maintainers of their maintainership ... maybe that will change now though since Trump appears to be pro-Russia) | 17:56 |
| arno11 | lol indeed | 17:57 |
| Wizzup | yeah I wouldn't want to move too much to different services | 18:06 |
| freemangordon | Wizzup: 2:2.24.33-2+m7.8 | 18:08 |
| freemangordon | this is the version in daedalus | 18:08 |
| freemangordon | it should be 2:2.24.33-2+4m7.8, no? | 18:08 |
| Wizzup | or gtk? | 18:10 |
| Wizzup | yes, agreed, that needs to be fixed | 18:10 |
| freemangordon | yes, gtk | 18:10 |
| Wizzup | will rebuild | 18:17 |
| freemangordon | thanks | 18:20 |
| Wizzup | freemangordon: done | 19:39 |
| arno11 | got my first HAM/apt-worker notif on daedalus. works fine, no slowness | 19:59 |
| arno11 | are meta pkgs updates actually safe or should i need to wait a bit ? | 20:00 |
| arno11 | (seems all 4m7) | 20:01 |
| Wizzup | should be ok I think | 20:23 |
| arno11 | ok will try in a bit | 20:43 |
| dsc_ | i will veto gitlab | 21:11 |
| dsc_ | just so you know lol | 21:11 |
| arno11 | Wizzup: meta pkgs updates from HAM work fine but a bit bewildering as it works exactly like in fremantle: it disconnects everything included H-D, making the device unresponsive for a while (progress bar works fine during the update) | 21:24 |
| arno11 | then it reboots automatically | 21:25 |
| arno11 | showing the success update msg once in H-D | 21:25 |
| arno11 | btw i tried to update qt-input-method for example and now unmet deps errors disappeared | 21:29 |
| arno11 | so it seems ok | 21:30 |
| arno11 | hmm if i dist-upgrade, it wants to update hildon-meta-core but wants to downgrade some libgtk pkgs i have, and libclutter | 21:33 |
| arno11 | same with apt upgrade | 21:34 |
| arno11 | *libclutter-0.8-0 | 21:35 |
| arno11 | ah it says downgrade but wants to upgrade to 4m7 instead of m7.1 | 21:41 |
| arno11 | freemangordon: ^ | 21:41 |
| arno11 | so i suppose i should accept downgrade, no ? | 21:41 |
| arno11 | hmm really unclear...in ham i still have previous meta pkgs versions | 21:45 |
| arno11 | anyway, will upgrade from apt and see | 21:49 |
| arno11 | ah ok, for apt m7.1 > 4m7 | 21:55 |
| arno11 | from apt-cache policy, things seem ok now | 21:56 |
| arno11 | *after apt upgrade | 21:57 |
| arno11 | cool | 21:57 |
| Wizzup | arno11: yes, that was on purpose @ like in fremantle | 22:09 |
| Wizzup | arno11: accept the downgrades | 22:09 |
| Wizzup | we're still clearing up some mistake in the CI/CD where most chimaera packages were versioned wrong | 22:09 |
| Wizzup | anything +m7 is bad, it should have been +3m7 for chimaera | 22:10 |
| arno11 | yeah ok | 22:10 |
| Wizzup | daedalus is +4m7 - but that's 'lower' than +m7 | 22:10 |
| arno11 | understood | 22:10 |
| Wizzup | this is why we have these pin files to orce daedalus pkgs to be installed over chimaera | 22:10 |
| Wizzup | the gtk/libcal/cityinfo/libclutter were build with the wrong version for daedalus - +m7 | 22:10 |
| Wizzup | I only found out a bit into the build process and these are normally the packages we build first | 22:10 |
| Wizzup | I just fixed this today and that's why you are getting these 'DOWNGRADE' warnings, they will not be there upon the real release | 22:11 |
| Wizzup | part of my steps now is to find others like these and fix it | 22:11 |
| Wizzup | regarding the 'full system upgrade' ham upgrade pattern, that is intentional, but we can control this and decide not to do it. | 22:11 |
| arno11 | ok, thx for clarifications | 22:11 |
| Wizzup | I was amused it worked and decided to turn it on | 22:11 |
| Wizzup | and re qt-input-maemo, this is some CI/CD issue where things get messed up if you build them concurrently | 22:12 |
| Wizzup | which is what I think dsc normally does :) | 22:12 |
| arno11 | ok :) | 22:13 |
| dsc_ | who what where | 22:18 |
| dsc_ | im innocent :( | 22:18 |
| Wizzup | yeah it is a ci/cd problem | 22:19 |
| arno11 | dsc_: btw, any idea why conversations chat windows show the main window if i use 'force raster surface' on boot ? | 23:08 |
| arno11 | *on boot or at least if rasterisation starts before conversations | 23:10 |
| arno11 | sorry, scratch that, qt5ct creates a xsession.d file and modify some stuff before conversations starts. will try to move qt5ct after conversations | 23:17 |
| dsc_ | arno11: your warranty is void when using conversations in a manner that is not certified by Leste's Standards for Stability(c)(tm), such as pushing pixels over the CPU | 23:59 |
| dsc_ | arno11: that said, ill see if I can reproduce in a VM | 23:59 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!