Comments

40 Comments

Thanks for the testing so far! :) We are updating the Anaconda build with one that has some more fixes for F34.

karma

Fixes an issue with "fedpkg commit --with-changelog --message" not working on "rawhide" branch and also otherwise seems functional.

@pwhalen I've added a newer Initial Setup & deepndent Anaconda build with another IoT related bug fix. Could you test it once more ? Thanks in advance! :)

@rdieter No problem & sorry for not responding sooner. :)

And this is the Initial Setup Bodhi update with the fix:

FEDORA-2019-ade3788aac

That Initial Setup bug needs a fix in the Initial Setup codebase, so we can simply do a IS only update that fixes it once it reaches stable.

Note that Anaconda and libblockdev were added to this update as the FE 1637021 requires newer (>=3.6) DNF and the Anaconda build also contains FE 1635252 which needs newer libblockdev.

This update also needs python-blivet-3.1.0-2.fc29 to be added, as it was part of FEDORA-2018-85853d46b5 which for some reason newer got to compose and is now obsolete in regards to this bodhi update.

If this updates goes to stable without the updated Blivet, there could be issues with the resulting compose, especially on ARM.

This update has been unpushed.

This has been obsoleted by FEDORA-2018-f16a71bc92.

karma

Work good, thanks a lot & thanks a lot for the log filtering backport! :)

Yet again - can't edit the update for some reason (?!) - so please add anaconda-28.22.7-1 whoever can, thanks!

I've looked on more details on how the "anaconda", "anaconda-core", "anaconda-tui" and "anaconda-gui" packages interact and are used. It seems to be safe to make the "anaconda" package depend on "anaconda-install-env-deps".

That way any expectations that the "anaconda" package pulls in everything will still hold.

Initial Setup depends on "anaconda-tui" and/or "anaconda-gui", which pull in "anaconda-core" - there is no direct dependency on the "anaconda" package, so the new dependency should not cause issues.

Anaconda PR: https://github.com/rhinstaller/anaconda/pull/1432

BTW, the Anaconda test day is actually on Monday, not Tuesday: https://fedoraproject.org/wiki/Test_Day:2017-04-06_AnacondaBlivetGUI

So it would be really good to get this fix to this update & to composes as soon as possible, due to all the delays that happen with our current pipeline.

My impression from discussion with pbrobinson is that the anaconda-install-env-deps change really important for ARM & ARM image size and it's something they wanted for a long time to finally happen. Also an Anaconda testday is planed for Tuesday where this and other changes can be tested.

anaconda-28.22.6-1.fc28 has the %files fix but I can't edit the update for some reason (again).

The following packages should be added once the update is unlocked:

anaconda-28.22.5-1 initial-setup-0.3.57-1 python-simpleline-1.1-1.fc28

This fixes the following F28 Final blocker: https://bugzilla.redhat.com/show_bug.cgi?id=1557472

Initial Setup initial-setup-0.3.56-1 needs to be added to this update due to some Anaconda API changes affecting it. I'll do that once the update is no longer pending.

karma

I've tried on a TTY with:

libinput debug-events

Then connecting the Aiptek tablet and drawing on it with the pen. That always crashed libinput before with libinput-1.9.0-1.fc26.x86_64, but it no longer happens and I get a stream of events instead when the pan is drawing on the tablet.

I've hit the same file conflict with regular cura during package installation, but after removing cura and installing just cura-lulzbot the application itself seems to work fine.

I was able to open a model, slice it and the gcode is now being printed on our TAZ6 in the lab.

@dustymabe Thanks for spotting the issue! Should be fixed in anaconda-27.20.4-6.fc27. :)