Comments
kkofler
With calamares-3.2.8-3.fc30, Calamares should now finally be fully working again.
I removed #1639021 from the bugs claimed fixed, hopefully upstream can fix it for real soon.
This will never go to stable.
This update has been submitted for batched by jvanek.
This update has been submitted for stable by jvanek.
This update has been pushed to stable.
LOL, WTF?
This breaks system updates for all users with yum-utils installed: https://bugzilla.redhat.com/show_bug.cgi?id=1680221
(Workaround: dnf update --exclude=dnf-utils on the command line.)
That appears to be an old issue not caused by this update, so please report it to https://bugzilla.redhat.com/ , it is off-topic here!
Rkward still broken, see: https://bodhi.fedoraproject.org/updates/FEDORA-2018-b11015006b#comment-839064 https://bugzilla.redhat.com/show_bug.cgi?id=1598483#c6
Rkward must be upgraded to 0.7.0 (as was done on F28) to work with R 3.5.x.
FYI, this update broke Rkward! https://bugzilla.redhat.com/show_bug.cgi?id=1598483#c6
This update breaks QtWebEngine on a large set of machines on a primary architecture (all ARM machines without NEON) because the skia-neon patch is missing. It additionally also breaks QtWebEngine on a large set of machines on a secondary architecture (all i686 machines without SSE2) because the no-sse2 patch is missing. The no-sse2 patch is definitely hard to port, the skia-neon patch probably not that much. I never approved this version for anything other than Rawhide.
In addition, the QtWebEngine security fixes are not documented in the update's description and the update is not marked as a security update.
I think you should disable autopush also on this one (the F28 version), see the discussion on the F27 version.
The missing Provides are indeed fixed, nothing tries to drag in Java 9 anymore. Also no regressions noted.
(By the way, I don't think enabling autopush is a good idea here, because this update depends on the NSS update.)
I pulled this in as a dependency of the latest OpenJDK update. Everything on the system seems working as before, I don't see any regressions.
I can confirm that this fixes https://bugreports.qt.io/browse/QTBUG-68939 in both Kompare and KDevelop.
Everything else keeps working as before.
Unfortunately, this update breaks Kompare and KDevelop: https://bugreports.qt.io/browse/QTBUG-68939
Why did this not get any negative karma before getting pushed to stable?
Do you want me to file a downstream bug report too?
Less dependency bloat is always a good thing, ship it!
FYI, Qt 5 is included in RHEL 7, and it was recently upgraded there from 5.6 LTS to 5.9 LTS, is that still too old?
The issue I am having might actually not be caused by this update after all. I have also seen this come up on a F27 remix around the same time (~a week ago), and no grub2 update reached stable there at that time.
See: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/thread/F6QQ4EGTOBVDJ7HHUO7ZW3GEHGA7XQT5/
qt5-qtwebengine-freeworld is not part of Fedora and so when anything needs a version-locked depedency than qt5-qtwebengine-freeworld itself - there is no justification that a non-fedora package triggers a negative karma for a fedora package!
Please read my description of the issue. qt5-qtwebengine-freeworld correctly has a version-locked dependency on qt5-qtwebengine and qt5-qtbase, which is why qt5-qtbase is being held back in the first place. It is qt5-qtwebchannel that is missing the version-locked dependency on qt5-qtbase, which lets dnf upgrade qt5-qtwebchannel while holding back qt5-qtbase. This is an error in qt5-qtwebchannel packaging and must be fixed in qt5-qtwebchannel packaging.
Translations are back, thanks!