[Webkit-unassigned] [Bug 183346] [GTK] NetworkProcess from WebKitGtk+ 2.19.91 SIGSEVs in _FUN() at DerivedSources/ForwardingHeaders/wtf/Function.h:56

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=183346

--- Comment #9 from Michael Catanzaro <mcatanzaro at igalia.com> ---
It works fine for me. Problem is that Flatpak makes debugging much too hard. To debug this, we'll have to apply a test patch to the official GNOME Flatpak runtime, then wait a day for the build to complete, then test it, then try again if we need to try another debug patch.

There is an alternative: flapjack. It's not fun. There are no incremental builds in flapjack: every build is a full rebuild. But it has worked well for me in the past when needing to debug flatpak-specific issues. Setting up a flapjack build environment might be more work than Andres might want to do, but that seems like possibly the easiest patch forward. Andres, are you willing to try building WebKit with Flapjack, to see if you can reproduce there? (You would then call 'flapjack run org.gnome.Epiphany' to test the result.)

Alternatively, we could try to debug this on the keyring side of things. It's clear from the code that there is a problem with one particular secret on one particular website. So you should look with seahorse to see if there are any obvious problems with the secret (e.g. is it empty?). Make a copy of your login keyring somewhere safe, then delete the secret and see if the crash goes away.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20180307/64c9b99d/attachment.html>