Version-Release number of selected component: gnome-settings-daemon-3.18.2-1.fc24 Additional info: reporter: libreport-2.6.3 backtrace_rating: 4 cmdline: /usr/libexec/gnome-settings-daemon crash_function: CD_IS_DEVICE executable: /usr/libexec/gnome-settings-daemon global_pid: 2284 kernel: 4.4.0-0.rc3.git4.1.fc24.x86_64 runlevel: N 5 type: CCpp uid: 42 Truncated backtrace: Thread no. 1 (8 frames) #0 CD_IS_DEVICE at cd-device.h:41 #1 cd_device_dbus_properties_changed_cb at cd-device.c:624 #2 ffi_call_unix64 at ../src/x86/unix64.S:76 #3 ffi_call at ../src/x86/ffi64.c:525 #4 g_cclosure_marshal_generic at gclosure.c:1490 #9 on_properties_changed at gdbusproxy.c:1139 #10 emit_signal_instance_in_idle_cb at gdbusconnection.c:3701 #15 gtk_main at gtkmain.c:1241
Created attachment 1102794 [details] File: backtrace
Created attachment 1102795 [details] File: cgroup
Created attachment 1102796 [details] File: core_backtrace
Created attachment 1102797 [details] File: dso_list
Created attachment 1102798 [details] File: environ
Created attachment 1102799 [details] File: exploitable
Created attachment 1102800 [details] File: limits
Created attachment 1102801 [details] File: maps
Created attachment 1102802 [details] File: mountinfo
Created attachment 1102803 [details] File: namespaces
Created attachment 1102804 [details] File: open_fds
Created attachment 1102805 [details] File: proc_pid_status
Created attachment 1102806 [details] File: var_log_messages
also happens with 20151219 Workstation rawhide build with gnome-settings-daemon-3.19.4.1.fc24.x86_64.
I'm really struggling to work out what's going on here; we disconnect the signals on finalize so there must be some kind of race. Can you give me any hints on how you might be able to trigger this? Thanks.
Boot any of the Rawhide nightlies, and when the desktop comes up, I get a notification that gnome-settings-daemon has crashed. Always happens.
Similar problem has been detected: Not sure reporter: libreport-2.6.4 backtrace_rating: 4 cmdline: /usr/libexec/gnome-settings-daemon crash_function: CD_IS_DEVICE executable: /usr/libexec/gnome-settings-daemon global_pid: 2152 kernel: 4.5.0-0.rc2.git2.1.fc24.x86_64 package: gnome-settings-daemon-3.19.5-1.fc24 reason: gnome-settings-daemon killed by SIGSEGV runlevel: N 5 type: CCpp uid: 42
*** Bug 1306014 has been marked as a duplicate of this bug. ***
*** Bug 1306461 has been marked as a duplicate of this bug. ***
Similar problem has been detected: Boot into Fedora-Workstation-Live-x86_64-rawhide-20160220.iso with enforcing=0 to avoid SELinux issues. reporter: libreport-2.6.4 backtrace_rating: 4 cmdline: /usr/libexec/gnome-settings-daemon crash_function: CD_IS_DEVICE executable: /usr/libexec/gnome-settings-daemon global_pid: 1899 kernel: 4.5.0-0.rc4.git3.1.fc24.x86_64 package: gnome-settings-daemon-3.19.90-1.fc24 reason: gnome-settings-daemon killed by SIGSEGV runlevel: unknown type: CCpp uid: 1000
*** Bug 1311026 has been marked as a duplicate of this bug. ***
This bug appears to have been reported against 'rawhide' during the Fedora 24 development cycle. Changing version to '24'. More information and reason for this action is here: https://fedoraproject.org/wiki/Fedora_Program_Management/HouseKeeping/Fedora24#Rawhide_Rebase
Is this the same crash? #0 0x00007f2d27364fab in cd_device_dbus_properties_changed_cb () at /lib64/libcolord.so.2 #1 0x00007f2d5895cc58 in ffi_call_unix64 () at /lib64/libffi.so.6 #2 0x00007f2d5895c6ba in ffi_call () at /lib64/libffi.so.6 #3 0x00007f2d59470d9e in g_cclosure_marshal_generic () at /lib64/libgobject-2.0.so.0 #4 0x00007f2d59470578 in g_closure_invoke () at /lib64/libgobject-2.0.so.0 #5 0x00007f2d59483152 in signal_emit_unlocked_R () at /lib64/libgobject-2.0.so.0 #6 0x00007f2d5948c45c in g_signal_emit_valist () at /lib64/libgobject-2.0.so.0 #7 0x00007f2d5948c82f in g_signal_emit () at /lib64/libgobject-2.0.so.0 #8 0x00007f2d5979c0b9 in on_properties_changed () at /lib64/libgio-2.0.so.0 #9 0x00007f2d5978b367 in emit_signal_instance_in_idle_cb () at /lib64/libgio-2.0.so.0 #10 0x00007f2d5919bba3 in g_main_context_dispatch () at /lib64/libglib-2.0.so.0 #11 0x00007f2d5919bf50 in g_main_context_iterate.isra () at /lib64/libglib-2.0.so.0 #12 0x00007f2d5919c272 in g_main_loop_run () at /lib64/libglib-2.0.so.0 #13 0x00007f2d5ad43735 in gtk_main () at /lib64/libgtk-3.so.0 #14 0x000056152596467b in main () gnome-settings-daemon-3.19.90-1.fc24.x86_64 colord-1.3.1-3.fc24.x86_64 colord-libs-1.3.1-3.fc24.x86_64
Just briefly looking at the code, gsd_color_manager_start calls gsd_color_state_start, which starts all sort of async stuff, while operating on a borrowed reference to the state object that was passed in from the color manager. gsd_color_manager_stop does nothing to interrup the async stuff that might be happening with the state object. and finalize throws away the reference to the state object. That seems like an obvious opportunity for the async machinery to operate on a dead state object. Not saying this is what is happening here, but
Filed downstream at https://bugzilla.gnome.org/show_bug.cgi?id=763382
*** Bug 1316850 has been marked as a duplicate of this bug. ***
*** Bug 1317199 has been marked as a duplicate of this bug. ***
Similar problem has been detected: Bug reproduced when gnome starts. reporter: libreport-2.6.4 backtrace_rating: 4 cmdline: /usr/libexec/gnome-settings-daemon crash_function: CD_IS_DEVICE executable: /usr/libexec/gnome-settings-daemon global_pid: 2040 kernel: 4.5.0-0.rc7.git3.1.fc25.x86_64 package: gnome-settings-daemon-3.19.91-1.fc25 reason: gnome-settings-daemon killed by SIGSEGV runlevel: N 5 type: CCpp uid: 1000
Similar problem has been detected: I booted the computer and logged into a GNOME Session (Xorg). This has also happened in a Wayland session previously. reporter: libreport-2.6.4 backtrace_rating: 4 cmdline: /usr/libexec/gnome-settings-daemon crash_function: CD_IS_DEVICE executable: /usr/libexec/gnome-settings-daemon global_pid: 2496 kernel: 4.5.0-0.rc7.git0.2.fc24.x86_64 package: gnome-settings-daemon-3.19.91-1.fc24 reason: gnome-settings-daemon killed by SIGSEGV runlevel: N 5 type: CCpp uid: 42
Issue still present for gnome-settings-daemon-3.19.92-1.fc24.
(In reply to Joachim Frieben from comment #30) > Issue still present for gnome-settings-daemon-3.19.92-1.fc24. If you're looking into the upstream bug report and git repos [1] you'll see that this was fixed right after 3.19.92 was released. [1] https://git.gnome.org/browse/gnome-settings-daemon/log/
Proposed as a Blocker for 24-beta by Fedora user genodeftest using the blocker tracking app because: Any custom (user) settings for both Gnome user sessions and GDM sessions queried from gnome-settings-daemon are ignored as a result of g-s-d crashing immediately after being started. This violates: Beta release criterium: Desktop panel No part of any release-blocking desktop's panel (or equivalent) configuration may crash on startup or be entirely non-functional. Final release criterium: Default panel functionality All elements of the default panel (or equivalent) configuration in all release-blocking desktops must function correctly in typical use. PS: I haven't proposed any blocker or freeze exception before. I might have done it wrong. If I did, please tell me.
Discussed at today's blocker review meeting [1]. Voted as AcceptedBlocker (Beta) - this is clearly a serious issue (g-s-d crashing has all sorts of consequences). It's not entirely clear exactly what triggers it, but certainly multiple people have run into it, so the impact seems severe enough to accept as a blocker [1] https://meetbot-raw.fedoraproject.org/fedora-blocker-review/2016-03-21
Similar problem has been detected: bare metal install/ user and root set in installer reporter: libreport-2.6.4 backtrace_rating: 4 cmdline: /usr/libexec/gnome-settings-daemon crash_function: CD_IS_DEVICE executable: /usr/libexec/gnome-settings-daemon global_pid: 2122 kernel: 4.5.0-0.rc7.git0.2.fc24.x86_64 package: gnome-settings-daemon-3.19.91-1.fc24 reason: gnome-settings-daemon killed by SIGSEGV runlevel: N 5 type: CCpp uid: 1000
Similar problem has been detected: Boot F24 Alpha candidate 7 (Alpha-1.7) Workstation x86_64 live on bare metal. Crash notification appeared as soon as the desktop came up. reporter: libreport-2.6.4 backtrace_rating: 4 cmdline: /usr/libexec/gnome-settings-daemon crash_function: CD_IS_DEVICE executable: /usr/libexec/gnome-settings-daemon global_pid: 1954 kernel: 4.5.0-0.rc7.git0.2.fc24.x86_64 package: gnome-settings-daemon-3.19.91-1.fc24 reason: gnome-settings-daemon killed by SIGSEGV runlevel: unknown type: CCpp uid: 1000
There's a 3.20.0 release that should go in an update shortly: http://koji.fedoraproject.org/koji/buildinfo?buildID=747442 If rel-eng wants to drag that into the compose, I'll let them do the necessary steps.
Nah, no need, we'll just pick it up for Beta. I was just filing everything I saw in Alpha testing.
Similar problem has been detected: Just load Fedora Workstation 24 Alpha from live USB reporter: libreport-2.6.4 backtrace_rating: 4 cmdline: /usr/libexec/gnome-settings-daemon crash_function: CD_IS_DEVICE executable: /usr/libexec/gnome-settings-daemon global_pid: 2113 kernel: 4.5.0-0.rc7.git0.2.fc24.x86_64 package: gnome-settings-daemon-3.19.91-1.fc24 reason: gnome-settings-daemon killed by SIGSEGV runlevel: unknown type: CCpp uid: 1000
*** Bug 1322243 has been marked as a duplicate of this bug. ***
Similar problem has been detected: happening at startup on a live DVD reporter: libreport-2.6.4 backtrace_rating: 4 cmdline: /usr/libexec/gnome-settings-daemon crash_function: CD_IS_DEVICE executable: /usr/libexec/gnome-settings-daemon global_pid: 1772 kernel: 4.5.0-0.rc7.git0.2.fc24.x86_64 package: gnome-settings-daemon-3.19.91-1.fc24 reason: gnome-settings-daemon killed by SIGSEGV runlevel: N 5 type: CCpp uid: 1000
*** Bug 1322631 has been marked as a duplicate of this bug. ***
*** Bug 1322694 has been marked as a duplicate of this bug. ***
Similar problem has been detected: just load Fedora from USB live flashdrive reporter: libreport-2.6.4 backtrace_rating: 4 cmdline: /usr/libexec/gnome-settings-daemon crash_function: CD_IS_DEVICE executable: /usr/libexec/gnome-settings-daemon global_pid: 1828 kernel: 4.5.0-0.rc7.git0.2.fc24.x86_64 package: gnome-settings-daemon-3.19.91-1.fc24 reason: gnome-settings-daemon killed by SIGSEGV runlevel: unknown type: CCpp uid: 1000
*** Bug 1322972 has been marked as a duplicate of this bug. ***
Similar problem has been detected: Just boot Fedora Live USB reporter: libreport-2.6.4 backtrace_rating: 4 cmdline: /usr/libexec/gnome-settings-daemon crash_function: CD_IS_DEVICE executable: /usr/libexec/gnome-settings-daemon global_pid: 1810 kernel: 4.5.0-0.rc7.git0.2.fc24.x86_64 package: gnome-settings-daemon-3.19.91-1.fc24 reason: gnome-settings-daemon killed by SIGSEGV runlevel: unknown type: CCpp uid: 1000
This bug has not appeared lately in installed systems with Fedora 23 including package gnome-settings-daemon-3.18.3-1.fc23 and with Fedora 24 including package gnome-settings-daemon-3.20.0-1.fc24.
*** Bug 1323565 has been marked as a duplicate of this bug. ***
*** Bug 1327841 has been marked as a duplicate of this bug. ***
*** Bug 1327842 has been marked as a duplicate of this bug. ***
*** Bug 1329021 has been marked as a duplicate of this bug. ***
*** Bug 1329826 has been marked as a duplicate of this bug. ***
*** Bug 1331929 has been marked as a duplicate of this bug. ***
*** Bug 1331962 has been marked as a duplicate of this bug. ***