Bug 2363595 - [abrt] evolution: WebKit::waitUntilSyncedOrDie(): evolution killed by SIGTRAP
Summary: [abrt] evolution: WebKit::waitUntilSyncedOrDie(): evolution killed by SIGTRAP
Keywords:
Status: CLOSED DUPLICATE of bug 2162351
Alias: None
Product: Fedora
Classification: Fedora
Component: webkitgtk
Version: 42
Hardware: x86_64
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
Assignee: Milan Crha
QA Contact: Fedora Extras Quality Assurance
URL: https://retrace.fedoraproject.org/faf...
Whiteboard: abrt_hash:df3e4e8ef72412f138a9e951ce3...
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2025-05-01 22:21 UTC by Davide Repetto
Modified: 2025-05-14 15:03 UTC (History)
9 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2025-05-05 08:54:28 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)
File: proc_pid_status (1.51 KB, text/plain)
2025-05-01 22:21 UTC, Davide Repetto
no flags Details
File: maps (3.97 KB, text/plain)
2025-05-01 22:21 UTC, Davide Repetto
no flags Details
File: limits (1.29 KB, text/plain)
2025-05-01 22:21 UTC, Davide Repetto
no flags Details
File: environ (4.26 KB, text/plain)
2025-05-01 22:21 UTC, Davide Repetto
no flags Details
File: open_fds (6.71 KB, text/plain)
2025-05-01 22:21 UTC, Davide Repetto
no flags Details
File: mountinfo (3.91 KB, text/plain)
2025-05-01 22:21 UTC, Davide Repetto
no flags Details
File: os_info (729 bytes, text/plain)
2025-05-01 22:21 UTC, Davide Repetto
no flags Details
File: cpuinfo (2.43 KB, text/plain)
2025-05-01 22:21 UTC, Davide Repetto
no flags Details
File: core_backtrace (98.93 KB, text/plain)
2025-05-01 22:21 UTC, Davide Repetto
no flags Details
File: dso_list (472 bytes, text/plain)
2025-05-01 22:21 UTC, Davide Repetto
no flags Details
File: backtrace (226.43 KB, text/plain)
2025-05-01 22:21 UTC, Davide Repetto
no flags Details

Description Davide Repetto 2025-05-01 22:21:03 UTC
Description of problem:
When I clicked on the "new" message button, cpu hit 100% for a while, then Evolution crashed.

After that Evolution crashes on start with this dump.

Version-Release number of selected component:
evolution-3.56.1-1.fc42

Additional info:
reporter:       libreport-2.17.15
type:           CCpp
reason:         evolution killed by SIGTRAP
journald_cursor: s=0ca15e42f6714a5baadcee0db98ae15f;i=29df1;b=08f55ab501f5443099eca06974c245e9;m=1cbe2dbda6;t=63419d6330784;x=998ff1877e2b6e8d
executable:     /usr/bin/evolution
cmdline:        evolution
cgroup:         0::/user.slice/user-1000.slice/session-7.scope
rootdir:        /
uid:            1000
kernel:         6.14.4-300.fc42.x86_64
package:        evolution-3.56.1-1.fc42
runlevel:       N 5
backtrace_rating: 4
crash_function: WebKit::waitUntilSyncedOrDie

Truncated backtrace:
Thread no. 1 (57 frames)
 #4 WebKit::waitUntilSyncedOrDie at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/Launcher/glib/XDGDBusProxy.cpp:175
 #5 WebKit::XDGDBusProxy::launch at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/Launcher/glib/XDGDBusProxy.cpp:244
 #6 WebKit::bubblewrapSpawn at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/Launcher/glib/BubblewrapLauncher.cpp:910
 #7 WebKit::ProcessLauncher::launchProcess at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/Launcher/glib/ProcessLauncherGLib.cpp:238
 #8 WebKit::ProcessLauncher::ProcessLauncher at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/Launcher/ProcessLauncher.cpp:45
 #9 WebKit::ProcessLauncher::create at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/Launcher/ProcessLauncher.h:147
 #10 WebKit::AuxiliaryProcessProxy::connect at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/AuxiliaryProcessProxy.cpp:186
 #11 WebKit::WebProcessProxy::create at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/WebProcessProxy.cpp:279
 #12 WebKit::WebProcessPool::createNewWebProcess at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/WebProcessPool.cpp:796
 #13 WebKit::WebProcessPool::processForSite at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/WebProcessPool.cpp:1220
 #14 WebKit::WebPageProxy::launchProcess at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/WebPageProxy.cpp:1330
 #15 WebKit::WebPageProxy::launchInitialProcessIfNecessary at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/WebPageProxy.cpp:6157
 #16 WebKit::WebPageProxy::runJavaScriptInFrameInScriptWorld at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/WebPageProxy.cpp:6168
 #17 webkitWebViewRunJavaScriptWithParams at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/API/glib/WebKitWebView.cpp:4180
 #18 webkitWebViewEvaluateJavascriptInternal at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/API/glib/WebKitWebView.cpp:4239
 #19 webkit_web_view_run_javascript at /usr/src/debug/webkitgtk-2.48.1-2.fc42.x86_64/Source/WebKit/UIProcess/API/glib/WebKitWebView.cpp:4499
 #20 e_web_view_jsc_run_script at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/e-util/e-web-view-jsc-utils.c:308
 #21 e_web_view_jsc_add_rule_into_style_sheet at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/e-util/e-web-view-jsc-utils.c:459
 #22 e_web_view_update_styles at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/e-util/e-web-view.c:743
 #23 e_web_view_update_fonts_settings at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/e-util/e-web-view.c:3694
 #24 web_view_initialize at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/e-util/e-web-view.c:1369
 #25 web_view_constructed at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/e-util/e-web-view.c:1839
 #26 mail_display_constructed at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/mail/e-mail-display.c:1832
 #27 g_object_new_with_custom_constructor at ../gobject/gobject.c:2603
 #28 g_object_new_internal at ../gobject/gobject.c:2627
 #29 g_object_new_valist at ../gobject/gobject.c:2968
 #31 mail_paned_view_constructed at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/mail/e-mail-paned-view.c:790
 #32 g_object_new_internal at ../gobject/gobject.c:2680
 #34 g_object_new_valist at ../gobject/gobject.c:2968
 #36 e_mail_paned_view_new at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/mail/e-mail-paned-view.c:1302
 #37 mail_shell_content_constructed at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/modules/mail/e-mail-shell-content.c:176
 #38 g_object_new_internal at ../gobject/gobject.c:2680
 #40 g_object_new_valist at ../gobject/gobject.c:2968
 #42 e_mail_shell_content_new at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/modules/mail/e-mail-shell-content.c:351
 #43 mail_shell_view_constructed at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/modules/mail/e-mail-shell-view.c:589
 #44 g_object_new_internal at ../gobject/gobject.c:2680
 #46 g_object_new_valist at ../gobject/gobject.c:2968
 #48 shell_window_create_shell_view at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/shell/e-shell-window.c:388
 #49 e_shell_window_get_shell_view at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/shell/e-shell-window.c:892
 #50 e_shell_window_set_active_view at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/shell/e-shell-window.c:1096
 #52 object_set_property at ../gobject/gobject.c:2193
 #53 g_object_setv at ../gobject/gobject.c:3106
 #55 g_object_set_property at ../gobject/gobject.c:3406
 #56 g_settings_binding_key_changed at ../gio/gsettings.c:2781
 #57 g_settings_bind_with_mapping at ../gio/gsettings.c:3092
 #58 g_settings_bind at ../gio/gsettings.c:2916
 #59 e_shell_window_private_constructed at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/shell/e-shell-window-private.c:233
 #60 shell_window_constructed at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/shell/e-shell-window.c:261
 #61 g_object_new_internal at ../gobject/gobject.c:2680
 #63 g_object_new_valist at ../gobject/gobject.c:2968
 #65 e_shell_window_new at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/shell/e-shell-window.c:821
 #66 e_shell_create_shell_window at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/shell/e-shell.c:2913
 #67 handle_options_idle_cb at /usr/src/debug/evolution-3.56.1-1.fc42.x86_64/src/shell/e-shell.c:2066
 #70 g_main_context_dispatch_unlocked at ../glib/gmain.c:4249
 #71 g_main_context_iterate_unlocked at ../glib/gmain.c:4314
 #72 g_main_context_iteration at ../glib/gmain.c:4379
 #73 g_application_run at ../gio/gapplication.c:2715

Comment 1 Davide Repetto 2025-05-01 22:21:08 UTC
Created attachment 2088051 [details]
File: proc_pid_status

Comment 2 Davide Repetto 2025-05-01 22:21:09 UTC
Created attachment 2088052 [details]
File: maps

Comment 3 Davide Repetto 2025-05-01 22:21:10 UTC
Created attachment 2088053 [details]
File: limits

Comment 4 Davide Repetto 2025-05-01 22:21:12 UTC
Created attachment 2088054 [details]
File: environ

Comment 5 Davide Repetto 2025-05-01 22:21:14 UTC
Created attachment 2088055 [details]
File: open_fds

Comment 6 Davide Repetto 2025-05-01 22:21:15 UTC
Created attachment 2088056 [details]
File: mountinfo

Comment 7 Davide Repetto 2025-05-01 22:21:16 UTC
Created attachment 2088057 [details]
File: os_info

Comment 8 Davide Repetto 2025-05-01 22:21:18 UTC
Created attachment 2088058 [details]
File: cpuinfo

Comment 9 Davide Repetto 2025-05-01 22:21:19 UTC
Created attachment 2088059 [details]
File: core_backtrace

Comment 10 Davide Repetto 2025-05-01 22:21:20 UTC
Created attachment 2088060 [details]
File: dso_list

Comment 11 Davide Repetto 2025-05-01 22:21:22 UTC
Created attachment 2088061 [details]
File: backtrace

Comment 12 Davide Repetto 2025-05-01 22:22:33 UTC
I clicked on the "new" message button, cpu hit 100% for a while, then Evolution crashed.

Now Evolution refuses to start as in:

:~$ evolution
bwrap: Can't find source path /root/.cache/at-spi: Permission denied

** (org.gnome.Evolution:490827): ERROR **: 00:03:45.754: Failed to fully launch dbus-proxy: Processo figlio uscito con codice 1
Rilevato trace/breakpoint (core dump creato)

For the second crash I opened bug 2363595.


reporter:       libreport-2.17.15
type:           CCpp
reason:         evolution killed by SIGTRAP
journald_cursor: s=0ca15e42f6714a5baadcee0db98ae15f;i=29de6;b=08f55ab501f5443099eca06974c245e9;m=1cb0397230;t=63419c83ebc0e;x=ec12f277a3d5a9b0
executable:     /usr/bin/evolution
cmdline:        evolution
cgroup:         0::/user.slice/user-1000.slice/session-7.scope
rootdir:        /
uid:            1000
kernel:         6.14.4-300.fc42.x86_64
package:        evolution-3.56.1-1.fc42
runlevel:       N 5
backtrace_rating: 4
crash_function: WebKit::waitUntilSyncedOrDie

Comment 13 Davide Repetto 2025-05-01 22:29:08 UTC
note:
Evolution has been giving me the warning "bwrap: Can't find source path /root/.cache/at-spi: Permission denied", bu I never run it a root.

Comment 14 Davide Repetto 2025-05-01 22:29:54 UTC
note:
Evolution has been giving me the warning "bwrap: Can't find source path /root/.cache/at-spi: Permission denied", but I never run it a root.

Comment 15 Milan Crha 2025-05-05 08:54:28 UTC
Thanks for a bug report. The at-spi problem is already logged as bug #2162351, thus I'll mark this a a duplicate of it.

There had been some investigation of this in a different distro [1]. The thing is that the display manager like gdm or lightdm runs the AT-SPI D-Bus daemon as the root. That daemon is meant to be closed once the user logs into the desktop environment, but it sometimes fails. If I recall correctly, the "bad" daemon is stopped few minutes after log in, then the WebKitGTK can run with no problem.

[1] https://forums.gentoo.org/viewtopic-p-8854461.html

*** This bug has been marked as a duplicate of bug 2162351 ***

Comment 16 Davide Repetto 2025-05-05 12:50:53 UTC
Thank, Milan. Unfortunately my note in comment 13 was not exactly clear:
My intention was to focus away from at-spi, because, actually, I _always_ get that error, even when evolution works perfectly. :)

In this case there must have been some kind of data corruption behind the crash, because the only way I could make it work again was to restore the whole .local/share/evolution/.

Thanks again.

Comment 17 Milan Crha 2025-05-05 13:22:52 UTC
> the only way I could make it work again was to restore the whole .local/share/evolution/.

That's odd, the ~/.local/share/evolution/ should not play a role in this. I guess the faulty leftover AT-SPI D-Bus service closed itself before the restore finished, then the evolution could start without any issue.

When I boot Fedora Workstation and stay in gdm, then there are running at-spi processes:

   3161 tty1     Sl+    0:00 /usr/libexec/at-spi-bus-launcher
   3168 tty1     S+     0:00 /usr/bin/dbus-daemon --config-file=/usr/share/defaults/at-spi2/accessibility.conf --nofork --print-address 12 --address=unix:path=/run/user/42/at-spi/bus
   3171 tty1     Sl+    0:00 /usr/libexec/at-spi2-registryd --use-gnome-session

and after I log in to the GNOME, these are replaced with:

   3956 ?        Ssl    0:00 /usr/libexec/at-spi-bus-launcher
   3963 ?        S      0:00 /usr/bin/dbus-broker-launch --config-file=/usr/share/defaults/at-spi2/accessibility.conf --scope user
   3965 ?        Ssl    0:00 /usr/libexec/at-spi2-registryd --use-gnome-session

I think the processes started in the gdm did not close quickly enough to be replaced by those sstarted in the user session in your case, which causes the problem for the WebKitGTK.

Comment 18 Milan Crha 2025-05-14 12:58:37 UTC
The Gentoo forum has a workaround:
https://forums.gentoo.org/viewtopic-p-8861791.html#8861791

Either run Firefox first or run a gdbus command mentioned there before the Evolution.

Michael, I thought it (the gdbus command) is what the WebKitGTK also does, only in the code, not in the command line. Or can it be you call that interface without a "can-autostart" flag?

Comment 19 Michael Catanzaro 2025-05-14 15:03:56 UTC
The only place WebKit uses this interface is queryAccessibilityBusAddress in WebProcessPoolGLib.cpp. It does not use G_DBUS_CALL_FLAGS_NO_AUTO_START, so it should auto-start. But we already know the problem is caused because it's already started.

Our discussion here is likely to get lost. Keep in mind there is an upstream issue for this: https://gitlab.gnome.org/GNOME/at-spi2-core/-/issues/180


Note You need to log in before you can comment on or make changes to this bug.