RESOLVED DUPLICATE of bug 258831 260343
segfault in evolution during gmail add new account oauth2 request
https://bugs.webkit.org/show_bug.cgi?id=260343
Summary segfault in evolution during gmail add new account oauth2 request
JonW
Reported 2023-08-17 04:45:53 PDT
Created attachment 467306 [details] various debug information When adding a new gmail account the program locks up when the gmail oauth2 dialogue is presented. The initial "enter email" web-dialogue is fine but pressing on next bugs out. If the email box is blanked out then it presents the "enter an email or phone number" error message with no problems. If an invalid email is entered it also bugs out. The bug causes the web-dialogue to blank out and the url to clear when tabbing from evolution to another program and back (or clicking on another application if evolution is not full screen). The "cancel" button below the web-dialogue can be pressed and it closes the window. I have attached a file containing: inxi output. evolution command line debug output (including bug). active xorg.conf. joburnalctl segfault information. webkit information from within evolution (reports the same no matter if WEBKIT_DISABLE_COMPOSITING_MODE= is 1 or 0). Should be noted that copy to clipboard button doesn't! I should also note that when I had 64G of ram in this laptop it would hard crash (like just randomly power off and the uefi bios would report "0" memory even though linux would report 64G and memtest would report 64G and pass with no problems). The amd spec is for 64G(2*32); the HP spec is for 16G(2*8); I currently have 32G(2*16) and have had no crashing issues since (multiple months). Just to make sure this was not the problem I downgraded to 16G and still got the same bugs. The reports are with 32G.
Attachments
various debug information (19.16 KB, text/plain)
2023-08-17 04:45 PDT, JonW
no flags
coredump and bt full (24.71 KB, text/plain)
2023-08-17 07:56 PDT, JonW
no flags
bt full (3.00 KB, text/plain)
2023-08-17 11:06 PDT, JonW
no flags
Michael Catanzaro
Comment 1 2023-08-17 07:02:51 PDT
Hi, to investigate a crash we need to see a backtrace taken with gdb, with all required debuginfo installed. Please see these instructions: https://blogs.gnome.org/mcatanzaro/2021/09/18/creating-quality-backtraces-for-crash-reports/. This will point to exactly what is going wrong. Thanks! > Should be noted that copy to clipboard button doesn't! It doesn't work because Evolution disables JavaScript. I suppose we shouldn't show the button at all in that case. Feel free to report a separate bug for this.
JonW
Comment 2 2023-08-17 07:56:39 PDT
Created attachment 467308 [details] coredump and bt full
JonW
Comment 3 2023-08-17 07:59:45 PDT
(In reply to Michael Catanzaro from comment #1) > Hi, to investigate a crash we need to see a backtrace taken with gdb, with > all required debuginfo installed. Please see these instructions: > https://blogs.gnome.org/mcatanzaro/2021/09/18/creating-quality-backtraces- > for-crash-reports/. This will point to exactly what is going wrong. Thanks! Hi there, thanks for the reply. Have added the attachment. > > > Should be noted that copy to clipboard button doesn't! > > It doesn't work because Evolution disables JavaScript. I suppose we > shouldn't show the button at all in that case. Feel free to report a > separate bug for this.
Michael Catanzaro
Comment 4 2023-08-17 08:24:37 PDT
Hi, this backtrace is missing debuginfo. Please install the relevant debuginfo packages and try again. (Review the instructions in that blog post.)
Michael Catanzaro
Comment 5 2023-08-17 08:25:42 PDT
Actually, although we can only see the top frame, it matches bug #258831 so it's probably a duplicate of that one. *** This bug has been marked as a duplicate of bug 258831 ***
JonW
Comment 6 2023-08-17 11:06:53 PDT
JonW
Comment 7 2023-08-17 11:08:08 PDT
(In reply to Michael Catanzaro from comment #5) > Actually, although we can only see the top frame, it matches bug #258831 so > it's probably a duplicate of that one. > > *** This bug has been marked as a duplicate of bug 258831 *** I think I've managed to get the back trace to work now and have attached it for completeness.
Michael Catanzaro
Comment 8 2023-08-17 11:18:11 PDT
Good job. That backtrace is good and confirms that this is indeed the same as bug #258831. So you're not alone, at least.
Note You need to log in before you can comment on or make changes to this bug.