Comments
rathann
@emarci Yes, thanks for the reminder.
Breaks ABI compared to 0.99.6, please don't push without coordinating with all dependent packages.
Breaks ABI compared to 0.99.6, please don't push without coordinating with all dependent packages.
There's unannounced ABI breakage in this update:
$ abipkgdiff --d1 libfastjson-debuginfo-0.99.6-1.fc26.x86_64.rpm --d2 libfastjson-debuginfo-0.99.7-1.fc26.x86_64.rpm --devel1 libfastjson-devel-0.99.6-1.fc26.x86_64.rpm --devel2 libfastjson-devel-0.99.7-1.fc26.x86_64.rpm libfastjson-0.99.6-1.fc26.x86_64.rpm libfastjson-0.99.7-1.fc26.x86_64.rpm
================ changes of 'libfastjson.so.4.1.0'===============
Functions changes summary: 3 Removed, 1 Changed (57 filtered out), 4 Added functions
[...]
There should be NO removed and NO changed items. Please test each update for ABI changes in the future.
FYI this breaks rsyslog compiled with 0.99.7 but running with 0.99.6 installed.
Doesn't start after upgrade:
# rsyslogd -n -d
rsyslogd: symbol lookup error: rsyslogd: undefined symbol: fjson_global_do_case_sensitive_comparison
I did try 5.1.2 and indeed got the disappearing icon, so I'll hold off until 5.1.3. Thanks for the heads-up @robatino.
No need for a new bug report, thank you. I'm preparing an update to 5.1.2 as well. I'm glad 5.1.1 doesn't eat all your memory after all.
@michalvala If this were true, this update would've been marked as newpackage. It's marked as bugfix. Also, this is F27 update, not F26, so I don't know why you're talking about .fc26 package. Builds just for testing without the expectation they'll reach stable, should be released via COPR, not updates-testing.
@zzambers @michalvala Hold on, how can you give +1 karma when upgrading to this package is broken? This should have never landed in stable. It's not acceptable to break upgrades on purpose when there is a well-documented solution to avoid the problem.
Works for me.
The HTTPS Everywhere extension in F26 is not updated to WebExtension version and the maintainer wasn't planning to do this because that breaks SeaMonkey compatibility, see: FEDORA-2017-ecd295ec49 .
Thanks for the report. This doesn't happen on my machine. Was mozilla-noscript the only package you updated? Can you try reproducing with a clean user profile and only mozilla-noscript package installed (no other Firefox addons)? Which Firefox version? Please open a separate bug report if this is reproducible.
Why does this pull in python{,3}-gobject? It's not explained in either the commit message or in rpm changelog. Suddenly, I have cairo and a bunch of libraries related to graphical display installed on my headless server.
Why does this pull in python{,3}-gobject? It's not explained in either the commit message or in rpm changelog. Suddenly, I have cairo and a bunch of libraries related to graphical display installed on my headless server.
This broke FFmpeg-3.2.x, which doesn't support openjpeg 2.2 without patching.
The regressions from 4.11.11 are still unfixed: https://bugzilla.redhat.com/show_bug.cgi?id=1482649 https://bugzilla.redhat.com/show_bug.cgi?id=1483272 Works fine otherwise.
Works fine for me (tested using a Windows game).
Regressions since 4.11.11: 1. [i915] if external display is connected via HDMI, console output (esp. prompt for LUKS password) is no longer displayed on it (only on internal LCD). 2. [usb] keyboard and mouse input (connected via Logitech Unifying receiver, USB ID 046d:c52b) becomes erratic, hanging every few seconds and making the machine almost unusable until I disable USB autosuspend for the receiver USB device in powertop (I have powertop.service enabled).
It seems to work fine and the double tray icon issue is gone. However, the on/off switch doesn't seem to work at all, but that's a separate bug.