RikMills | mparillo: thanks :) | 06:56 |
---|---|---|
santa_ | hi everyone | 11:47 |
RikMills | hi | 11:48 |
santa_ | RikMills: yesterday I got a weird FTBFS in plasma-desktop, apparently something change in impish and that something broke the build | 11:48 |
santa_ | I could reproduce it in a PPA: https://launchpad.net/~panfaust/+archive/ubuntu/kde-test-bad/+packages | 11:49 |
santa_ | also in my server against fw 5.81 http://tritemio-groomlake.duckdns.org/build-status/buildstatus_ubuntu-exp2/ubuntu-exp2_status_plasma.html | 11:50 |
santa_ | and against fw 5.82 http://tritemio-groomlake.duckdns.org/build-status/buildstatus_ubuntu-exp2/ubuntu-exp2_status_plasma.html | 11:50 |
santa_ | previously it was building fine against both fw versions | 11:50 |
santa_ | I checked if they were cmake, pkg-config, xorg-server new versions in impish, but nope, they are the same versions we had before | 11:51 |
santa_ | if you look at the failure build logs: | 11:52 |
RikMills | -- The following features have been disabled: | 11:52 |
RikMills | XorgServer, XServer header needed for touchpad KCM (X11 backend) | 11:52 |
santa_ | 1. it fails to detect x server, yes | 11:52 |
santa_ | 2. missing files at the end of build log | 11:52 |
santa_ | I also checked the plasma-desktop git in case there was some kind of fix, but I found nothing | 11:53 |
santa_ | also note that in the PPA only fails for some archs :| that's weird | 11:54 |
RikMills | sometimes that means that whatever changed has only build for certain arches so far | 11:56 |
santa_ | yeah, maybe | 11:57 |
santa_ | -- Checking for module 'xorg-server' | 12:00 |
santa_ | -- Package 'valgrind', required by 'libdrm', not found | 12:00 |
santa_ | wat | 12:00 |
RikMills | so does it build with proposed disabled.... | 12:13 |
RikMills | libdrm build depends on valgrind [amd64 armhf i386 mips mipsel powerpc s390x] | 12:17 |
RikMills | which tallies with our failing arches | 12:18 |
RikMills | santa_: https://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html#libdrm | 12:20 |
RikMills | the fails there are failure to find valgrind on a build test | 12:20 |
BluesKaj | Hi folks | 12:30 |
RikMills | santa_: gotta go for a short while. back later | 12:33 |
santa_ | ack, I'm testing some things | 12:34 |
RikMills | builds ok without proposed | 12:34 |
santa_ | builds fine just installing valgrind, soe I guess addit it to build depends would workaround the thing | 13:04 |
santa_ | * adding | 13:08 |
santa_ | fails to build with nothing from proposed but libdrm-dev | 13:28 |
RikMills | ok, we can add that dep if and when we hit things | 13:28 |
RikMills | until it gets fixed | 13:29 |
santa_ | yep, I'm rebuilding in my PPA the failed build, I presume it might fail for all archs now | 13:30 |
RikMills | I tried that in a ci-train ppa, and it still failed | 13:31 |
RikMills | so I think on the arches that libdrm build deps on valgrind, it somehow now grows a dev package dependency on it | 13:33 |
RikMills | I am not sure why the new libdrm in propose might now do that | 13:33 |
RikMills | santa_: should note that riscv64 doesn't have a build of valgrind, so I guess any dep would need to be [!riscv64] | 13:36 |
RikMills | which I am trying now | 13:37 |
santa_ | I have the impression that libdrm-dev should depend on valgrind [amd64 armhf i386 mips mipsel powerpc s390x] | 13:37 |
santa_ | (thats what new libdrm has in Build-Depends | 13:38 |
santa_ | ) | 13:38 |
RikMills | agreed | 13:41 |
RikMills | or at least that seems logical on the facts | 13:41 |
RikMills | I think we are going to get a few things like this for this cycle, where ubuntu is syncing less well tested new things from experimental more often :/ | 13:44 |
santa_ | yeah, with sid freezed it's difficult to discern packages actually meant for debian experimental and packages which were uploaded to experimental because they can't be uploaded to sid | 14:12 |
santa_ | + the fact that debian experimental is not managed by britney which tends to reveal packaging problems | 14:12 |
santa_ | ok, so after a rebuild of plasma-desktop I can say you are right that the thing fails only for archs where valgrind is in libdrm build depends | 14:13 |
santa_ | http://tritemio-groomlake.duckdns.org/ka-iron-hand_reports/frameworks_archive/5.82_impish_retry_builds.pdf | 14:22 |
santa_ | http://tritemio-groomlake.duckdns.org/ka-iron-hand_reports/frameworks_archive/5.82_impish_proposed_migration.pdf | 14:22 |
RikMills | kool. thanks | 14:39 |
IrcsomeBot1 | <DarinMiller> TAG Everyone! | 14:51 |
IrcsomeBot1 | <DarinMiller> A question regarding usb-creator-kde. The 21.04 version seems to suffer from this bug: https://bazaar.launchpad.net/~usb-creator-hackers/usb-creator/trunk/revision/488. | 14:52 |
IrcsomeBot1 | <DarinMiller> If I manually install python3-sip, usb-creator-kde will run. But I am curious how the above patch should trickle down as a standard update. | 14:53 |
santa_ | probably that change will generate different depends | 14:54 |
santa_ | hmm, nope | 14:58 |
IrcsomeBot1 | <RikMills> @DarinMiller, 0.3.10 in updates should be fixed | 14:58 |
santa_ | Depends: ${misc:Depends}, ${python3:Depends}, | 14:58 |
santa_ | usb-creator-common (= ${binary:Version}), python3-pyqt5, python3-dbus.mainloop.pyqt5y | 14:58 |
IrcsomeBot1 | <RikMills> @DarinMiller 0.3.10 does not require python3-sip to be installed | 14:59 |
santa_ | depends of usb-creator-kde above, the stuff they are using after the change is probably provided by python3-pyqt5 | 15:00 |
santa_ | so indeed no need for python3-sip to be installed | 15:00 |
santa_ | (at least manually) | 15:02 |
IrcsomeBot1 | <RikMills> @DarinMiller the update to fix it in hirsute is 90% phased, so you may be in the 10% if you don't have it yet | 15:02 |
santa_ | https://launchpad.net/ubuntu/+source/usb-creator | 15:03 |
santa_ | still in hirsute-proposed | 15:04 |
santa_ | oh, and updates, nvm | 15:04 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!