On Sat, 2021-01-09 at 05:47 +0100, Axel Lieber wrote:
> Please attached the bt.txt file, without having installed any
> packages, as you/Milan suggested.
Hi,
thanks. The backtrace shows that almost all of the threads are in a
poll() call, waiting for orders. Two threads are related to WebKitGTK,
waiting in futex_...wait_cancelable(). I do not know whether it's
related to the composer creation. If there is anything reading the
attachment file or something, I do not see it in the backtrace. It can
run asynchronously.
I'm not sure how to move this further, I'm sorry. Running evolution
under valgrind/hellgrind or some such, but I doubt it would help in any
way.
Maybe, could you try this, please:
a) run evolution and select the message with which you can reproduce it
b) open a terminal and run this command:
$ ps ax | grep WebKitWeb
It will show opened WebKitWebProcess processes (or any with
a "similar" name)
c) switch to the Evolution and reproduce the problem
d) switch back to the terminal and re-run the `ps` command
e) if there is a new WebKitWebProcess, which was not there before, then
it's likely it's the one related to the composer window. Could you
grab a backtrace of it, please? It's the same gdb command as before,
only the --pid argument will have the process ID of the
WebKitWebProcess.
Also, is there a difference when you open a new composer (Ctrl+Shift+M)
and when you keep it open and try to reproduce the issue? When you keep
it open, the new composer window will re-use the existing
WebKitWebProcess.
Bye,
Milan
_______________________________________________
evolution-list mailing list
evolution-list@gnome.org
To change your list options or unsubscribe, visit ...
https://mail.gnome.org/mailman/listinfo/evolution-list
