Bug 1950838

Summary: LibreOffice Impress crashes with fcitx active
Product: [Fedora] Fedora Reporter: Leander <leander>
Component: fcitx-qt5Assignee: Qiyu Yan <yanqiyu01>
Status: CLOSED ERRATA QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: high Docs Contact:
Priority: unspecified    
Version: 34CC: i18n-bugs, i, liangsuilong, pwu, robinlee.sysu, yanqiyu01
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: fcitx-qt5-1.2.6-1.fc34 Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2021-05-02 02:02:37 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
/var/log/messages/ snippet around a crash
none
output from coredumpctl info none

Description Leander 2021-04-19 02:52:00 UTC
Created attachment 1773157 [details]
/var/log/messages/ snippet around a crash

Description of problem:

With fcitx enabled on FL34 LibreOffice Impress crashes 100% of the time. You don't even have to be using the program, the Impress window can be minimized and everything and just will die without error in the GUI. I upgraded this machine from FL33 and the problem did not exist there. I have fcitx-kkc kcm-fcitx installed on the KDE Plasma spin on an X11 session and this added to /etc/profile.d/fcitx.sh:

export XMODIFIERS="@im-fcitx"
export QT_IM_MODULE=fcitx
export GTK_IM_MODULE=fcitx

That file is executable, this setup has worked through several Fedora upgrades. 

TBH I'm not sure if fcitx is the problem here or if the bug is in LO or Plasma 5. Lots of moving parts. But if fcitx is completely exited (right click on the system tray and exit) LibreOffice seems stable. I'm attaching a snippet from /var/log/messages that seems to be relevant. 


Version-Release number of selected component (if applicable):

fcitx-libs-4.2.9.8-2.fc34.x86_64
fcitx-qt5-1.2.5-4.fc34.x86_64
fcitx-data-4.2.9.8-2.fc34.noarch
fcitx-gtk3-4.2.9.8-2.fc34.x86_64
fcitx-4.2.9.8-2.fc34.x86_64
fcitx-gtk2-4.2.9.8-2.fc34.x86_64
fcitx-kkc-0.1.4-2.fc34.x86_64
kcm-fcitx-0.5.6-3.fc34.x86_64

Operating System: Fedora 34
KDE Frameworks Version: 5.80.0
Qt Version: 5.15.2
Kernel Version: 5.11.14-300.fc34.x86_64
OS Type: 64-bit
Graphics Platform: X11

Version: 7.1.2.2
Build ID: 10(Build:2)
CPU threads: 6; OS: Linux 5.11; UI render: default; VCL: kf5
Locale: en-US (en_US.UTF-8); UI: en-US
Calc: threaded



How reproducible:

100% Every time.


Steps to Reproduce:
1. Enable fcitx
2. Open LO Impress
3. Use LO Impress (or let it idle with empty presentation
4. It spontaneously exists without and error

Actual results:

Dead LO Impress, nearly impossible to use with fcitx.

Expected results:

Work, as before. 

Additional info:

Comment 1 Qiyu Yan 2021-04-19 05:53:57 UTC
The log you attached is poorly formatted here, could you please run
coredumpctl info <PID of crashed process> 
to get a more clear traceback of the crash (I don't suggest directly uploading core file due to it may contain sensitive things)

And by saying "without error in the GUI", there isn't any record in ABRT?

Comment 2 Leander 2021-04-19 14:43:48 UTC
Created attachment 1773315 [details]
output from coredumpctl info

Comment 3 Leander 2021-04-19 14:44:37 UTC
(In reply to Qiyu Yan from comment #1)
> The log you attached is poorly formatted here, could you please run
> coredumpctl info <PID of crashed process> 
> to get a more clear traceback of the crash (I don't suggest directly
> uploading core file due to it may contain sensitive things)
> 
> And by saying "without error in the GUI", there isn't any record in ABRT?


Hello! Thank you for the follow up. 

Since I filed the report I realized ABRT wasn't installed. I can't remember if it didn't ship with the Plasma spin when I installed this machine's OS or it got removed at one point or the other. I installed it and the crash report does indeed show up in there and it will alert when it happens. But there is no indication of a problem from LibreOffice's GUI which makes sense since this appears to be a segfault.

My apologies on the log it was all I had to work with at the time I filed the report. I have uploaded an output from coredumpctl. Thanks!

Comment 4 Qiyu Yan 2021-04-24 04:25:59 UTC
Try this update, install and logout and relogin. to see if it fixes your problem
https://bodhi.fedoraproject.org/updates/FEDORA-2021-f356b2e66a

Comment 5 Fedora Update System 2021-04-24 14:33:17 UTC
FEDORA-2021-f356b2e66a has been submitted as an update to Fedora 34. https://bodhi.fedoraproject.org/updates/FEDORA-2021-f356b2e66a

Comment 6 Fedora Update System 2021-04-24 19:53:36 UTC
FEDORA-2021-f356b2e66a has been pushed to the Fedora 34 testing repository.
Soon you'll be able to install the update with the following command:
`sudo dnf upgrade --enablerepo=updates-testing --advisory=FEDORA-2021-f356b2e66a`
You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2021-f356b2e66a

See also https://fedoraproject.org/wiki/QA:Updates_Testing for more information on how to test updates.

Comment 7 Leander 2021-04-25 03:30:08 UTC
I installed the update and fcitx no longer causes LibreOffice Impress to through a segfault. Thanks!

Comment 8 Fedora Update System 2021-05-02 02:02:37 UTC
FEDORA-2021-f356b2e66a has been pushed to the Fedora 34 stable repository.
If problem still persists, please make note of it in this bug report.