1084597 - Firefox 33.0 freezes when waking up from sleep

Status: RESOLVED INVALID

(Firefox :: Untriaged, defect)

Description

[Simon Sapin (:SimonSapin)]

Since upgrading to 33.0, Firefox is unresponsive every time my laptop wakes from sleep-to-RAM. A few seconds after trying to interact with it, GNOME shows its "Firefox is unresponsive" dialog with the option to "Force quit", which seems to be the only way to recover. (Though I may have not tried waiting very long.)

Please let me know what I can do to help debug this. I’m running GNOME 3.14 (upgraded from 3.12 at the same time Firefox was upgraded from 32.x) on Archlinux x86_64.

Comment 1

[Simon Sapin (:SimonSapin)]

Thunderbird 32.2.0 seems affected too.

… or sometimes, at least.

Comment 2

[Matthias Versen [:Matti]]

What you could try is to find the regression range if you can reproduce it every time.
- http://mozilla.github.io/mozregression/

Comment 3

[Simon Sapin (:SimonSapin)]

I just assumed this issue was related to the upgrade to Firefox 33 since it started at the same time, but it might not be. (GNOME was updated at the same time.)

I now believe this is the issue: https://bugs.archlinux.org/task/42422

Firefox uses GTK+2, and Evince (GNOME’s PDF reader) uses GTK+3. Loading the Evince plugin into Firefox brings symbols for both GTK+ versions in the same process, which causes bad things to happen.

After removing libevbrowserplugin.so as suggested in the Archlinux bug tracker and making the laptop sleep and wake up, I can not reproduce the freeze which had happened every time so far.

I’ll close this in a while if I still can’t reproduce the issue. What is the correct resolution for a bug that is not in Firefox/Gecko but in a plugin? Invalid?

Comment 4

[:Gijs (he/him)]

(In reply to Simon Sapin (:SimonSapin) from comment #3)
> I’ll close this in a while if I still can’t reproduce the issue. What is the
> correct resolution for a bug that is not in Firefox/Gecko but in a plugin?
> Invalid?

Yup. Glad it seems to be fixed now!

Comment 5

[Simon Sapin (:SimonSapin)]

Yes, looks like https://bugs.archlinux.org/task/42422 was it.