qutebrowser / qutebrowser

A keyboard-driven, vim-like browser based on Python and Qt.
https://www.qutebrowser.org/
GNU General Public License v3.0
9.45k stars 1.01k forks source link

Renderer process crash when visiting atuin.sh website #8224

Closed ipvych closed 2 weeks ago

ipvych commented 2 weeks ago

Version info:

qutebrowser v3.1.0
Git commit: 
Backend: QtWebEngine 6.7, based on Chromium 118.0.5993.220 (from api)
Qt: 6.7.0

CPython: 3.11.9
PyQt: 6.7.0.dev2404081550

Qt wrapper info:
  PyQt6: success
  PyQt5: not imported
  -> selected: PyQt6 (via autoselect)

colorama: no
jinja2: 3.1.3
pygments: 2.17.2
yaml: 6.0.1
adblock: 0.6.0
objc: no
PyQt6.QtWebEngineCore: 6.6.0
PyQt6.sip: 6.7.12
pdf.js: 4.0.269 (/nix/store/pds26sdwnz0ifp3p828pphmcnw2vxbds-source/build/pdf.mjs)
sqlite: 3.45.2
QtNetwork SSL: OpenSSL 3.0.13 30 Jan 2024

Style: QFusionStyle
Platform plugin: xcb
OpenGL: AMD, 4.6 (Compatibility Profile) Mesa 24.0.5
Platform: Linux-6.8.6-zen1-x86_64-with-glibc2.39, 64bit
Linux distribution: NixOS 24.05 (Uakari) (nixos)
Frozen: False
Imported from /nix/store/3wmphq95vb3ziid419mlsipdba8l45mg-qutebrowser-3.1.0/lib/python3.11/site-packages/qutebrowser
Using Python from /nix/store/glfr70gi7hfaj50mwj2431p8bg60fhqw-python3-3.11.9/bin/python3.11
Qt library executable path: /nix/store/r8a6ss3xnc9j4yrgajbvy8wnn359cpkq-qtbase-6.7.0/libexec, data path: /nix/store/r8a6ss3xnc9j4yrgajbvy8wnn359cpkq-qtbase-6.7.0

Paths:
cache: /tmp/qutebrowser-basedir-9gx5_gbr/cache
config: /tmp/qutebrowser-basedir-9gx5_gbr/config
data: /tmp/qutebrowser-basedir-9gx5_gbr/data
runtime: /tmp/qutebrowser-basedir-9gx5_gbr/runtime
system data: /nix/store/3wmphq95vb3ziid419mlsipdba8l45mg-qutebrowser-3.1.0/share/qutebrowser

Autoconfig loaded: yes
Config.py: no config.py was loaded
Uptime: 0:00:01

Does the bug happen if you start with --temp-basedir?: Yes

Description Visiting the website causes render process crash, here is the log with debug toggle enabled, renderer process crash happens at 01:06:52 timestamp log.txt There is also a coredump produced when this crash happens, with result of running coredumpctl info <pid> here: dump_info.txt

How to reproduce Visit atuin.sh

The-Compiler commented 2 weeks ago

Relevant part:

[1610374:5:0607/010652.749143:FATAL:ffmpeg_glue.cc(160)] Check failed: container_ != container_names::MediaContainerName::kContainerUnknown (0 vs. 0)

stack:

#0 0x7404e5884ce5 base::debug::CollectStackTrace()
#1 0x7404e586cf26 base::debug::StackTrace::StackTrace()
#2 0x7404e578ce90 logging::LogMessage::~LogMessage()
#3 0x7404e578e631 logging::LogMessage::~LogMessage()
#4 0x7404e5793e4a logging::CheckError::~CheckError()
#5 0x7404e253d630 media::FFmpegGlue::OpenContext()
#6 0x7404e1547d86 base::internal::ReturnAsParamAdapter<>()
#7 0x7404e1547f4a base::internal::Invoker<>::RunOnce()
#8 0x7404e582febc base::internal::PostTaskAndReplyRelay::RunTaskAndPostReply()
#9 0x7404e582f9f5 base::internal::Invoker<>::RunOnce()
#10 0x7404e580b486 base::TaskAnnotator::RunTaskImpl()
#11 0x7404e583dfb3 base::TaskAnnotator::RunTask<>()
#12 0x7404e583e5cb base::internal::TaskTracker::RunSkipOnShutdown()
#13 0x7404e583e7d2 base::internal::TaskTracker::RunTask()
#14 0x7404e583dd38 base::internal::TaskTracker::RunAndPopNextTask()
#15 0x7404e584bf6a base::internal::WorkerThread::RunWorker()
#16 0x7404e584c74d base::internal::WorkerThread::RunPooledWorker()
#17 0x7404e584c8d0 base::internal::WorkerThread::ThreadMain()
#18 0x7404e5865c1a base::(anonymous namespace)::ThreadFunc()
#19 0x7404de0a1272 start_thread
#20 0x7404de11cc04 __GI___clone
Task trace:
#0 0x7404e25476c9 media::FFmpegDemuxer::Initialize()
#1 0x7404e2407d4f media::PipelineImpl::Start()
#2 0x7404ea31318b blink::MultiBufferDataSource::StartCallback()
#3 0x7404ea318f92 blink::MultiBufferReader::CheckWait()
#4 0x7404e60cc2b5 mojo::SimpleWatcher::Context::CallNotify()

I cannot reproduce. This might be something NixOS specific with how it builds QtWebEngine/ffmpeg. In any case, renderer process crashes typically aren't something qutebrowser can do anything about.

ipvych commented 2 weeks ago

Updating package to latest version fixed it, thanks for pointing out where the problem was, closing the issue then.