mate-panel --replace command crashes the panel
#1433
Comments
|
@mate-desktop/core-team ping |
|
Btw, na-tray and clock-applet and other applets will crash. |
|
I use mate-panel --replace all the time to restart the panel (e.g. in testing) and have not had this crash, Assuming you have the tray and clock out of process? If they were in-process they would take the panel down on crashing but you would not get a popup about either of them crashing |
|
All applets are build-in-process to support wayland in fedora. |
|
I also have 1.28 from the tarball installed and don't get the crash, so something is different between Fedora and Debian here. We could revert the offending commit and re-introduce it for 1.29 if we have to |
|
Let us wait with reverting until someone other can reproduce it or we get more reports. |
|
I cannot see any reason why that would fail, and if it's a mate-desktop issue I would expect a lot more problems than just this one |
|
This is what exactly happens. Ignore the EggSMClient-CRITICAL warning which spams the terminal too, when both commits are reverted.
I hope there isn't a problem with new mate image menu item api. |
|
I can't duplicate this. The double-free may depend on what exactly is in the panel or even the order in which things are loading, could be a race condition. If you can run the panel under valgrind AND if timing differences under valgrind don't block the crash you may be able to find it. EDIT: thinking the double-free might be something Valgrind can find |
|
I ran |
|
I don't know any way of running mate-panel --replace under gdb, as command line arguments seem to be ignored, giving |
|
@lukefromdc You might also want to adjust |
|
from: https://docs.fedoraproject.org/en-US/quick-docs/bugzilla-providing-a-stacktrace/ full stacktrace: https://www.dropbox.com/scl/fi/1g4thvu4qzc5xsme1ulqa/mate-panel_stacktrace?rlkey=93oefjgxiiobjnoi6h32wnsvb&dl=0 Same behaviour like without gdb. |
or inside gdb with the run command. |
|
Ok, hints to mate-panel are here: and |
|
Here's an AddressSanitizer report from Fedora 40: ReportERROR: AddressSanitizer: heap-use-after-free on address 0x5160000920c0 at pc 0x7ffff7f5b045 bp 0x7fffffffbda0 sp 0x7fffffffbd98
WRITE of size 8 at 0x5160000920c0 thread T0
#0 0x7ffff7f5b044 in mate_image_menu_item_finalize /home/s/code/mate-desktop/libmate-desktop/mate-image-menu-item.c:354:29
#1 0x7ffff70ad283 in g_object_unref (/lib64/libgobject-2.0.so.0+0x20283)
#2 0x7ffff7800dd4 (/lib64/libgtk-3.so.0+0x200dd4)
#3 0x7ffff7708272 (/lib64/libgtk-3.so.0+0x108272)
#4 0x7ffff709e43b in g_closure_invoke (/lib64/libgobject-2.0.so.0+0x1143b)
#5 0x7ffff70cde54 (/lib64/libgobject-2.0.so.0+0x40e54)
#6 0x7ffff70be653 (/lib64/libgobject-2.0.so.0+0x31653)
#7 0x7ffff70be8b0 in g_signal_emit_valist (/lib64/libgobject-2.0.so.0+0x318b0)
#8 0x7ffff70be972 in g_signal_emit (/lib64/libgobject-2.0.so.0+0x31972)
#9 0x7ffff79568e1 (/lib64/libgtk-3.so.0+0x3568e1)
#10 0x7ffff70ad439 in g_object_run_dispose (/lib64/libgobject-2.0.so.0+0x20439)
#11 0x7ffff7788567 (/lib64/libgtk-3.so.0+0x188567)
#12 0x7ffff7708272 (/lib64/libgtk-3.so.0+0x108272)
#13 0x7ffff709e43b in g_closure_invoke (/lib64/libgobject-2.0.so.0+0x1143b)
#14 0x7ffff70cde54 (/lib64/libgobject-2.0.so.0+0x40e54)
#15 0x7ffff70be653 (/lib64/libgobject-2.0.so.0+0x31653)
#16 0x7ffff70be8b0 in g_signal_emit_valist (/lib64/libgobject-2.0.so.0+0x318b0)
#17 0x7ffff70be972 in g_signal_emit (/lib64/libgobject-2.0.so.0+0x31972)
#18 0x7ffff79568e1 (/lib64/libgtk-3.so.0+0x3568e1)
#19 0x7ffff70ad439 in g_object_run_dispose (/lib64/libgobject-2.0.so.0+0x20439)
#20 0x7ffff7708272 (/lib64/libgtk-3.so.0+0x108272)
#21 0x7ffff709e43b in g_closure_invoke (/lib64/libgobject-2.0.so.0+0x1143b)
#22 0x7ffff70cde54 (/lib64/libgobject-2.0.so.0+0x40e54)
#23 0x7ffff70be653 (/lib64/libgobject-2.0.so.0+0x31653)
#24 0x7ffff70be8b0 in g_signal_emit_valist (/lib64/libgobject-2.0.so.0+0x318b0)
#25 0x7ffff70be972 in g_signal_emit (/lib64/libgobject-2.0.so.0+0x31972)
#26 0x7ffff79568e1 (/lib64/libgtk-3.so.0+0x3568e1)
#27 0x7ffff70ad439 in g_object_run_dispose (/lib64/libgobject-2.0.so.0+0x20439)
#28 0x7ffff77a7917 (/lib64/libgtk-3.so.0+0x1a7917)
#29 0x7ffff7708272 (/lib64/libgtk-3.so.0+0x108272)
#30 0x7ffff709e43b in g_closure_invoke (/lib64/libgobject-2.0.so.0+0x1143b)
#31 0x7ffff70cde54 (/lib64/libgobject-2.0.so.0+0x40e54)
#32 0x7ffff70be653 (/lib64/libgobject-2.0.so.0+0x31653)
#33 0x7ffff70be8b0 in g_signal_emit_valist (/lib64/libgobject-2.0.so.0+0x318b0)
#34 0x7ffff70be972 in g_signal_emit (/lib64/libgobject-2.0.so.0+0x31972)
#35 0x7ffff79568e1 (/lib64/libgtk-3.so.0+0x3568e1)
#36 0x7ffff70ad439 in g_object_run_dispose (/lib64/libgobject-2.0.so.0+0x20439)
#37 0x7ffff7967a70 (/lib64/libgtk-3.so.0+0x367a70)
#38 0x7ffff7708272 (/lib64/libgtk-3.so.0+0x108272)
#39 0x7ffff709e4d9 in g_closure_invoke (/lib64/libgobject-2.0.so.0+0x114d9)
#40 0x7ffff70cde54 (/lib64/libgobject-2.0.so.0+0x40e54)
#41 0x7ffff70be653 (/lib64/libgobject-2.0.so.0+0x31653)
#42 0x7ffff70be8b0 in g_signal_emit_valist (/lib64/libgobject-2.0.so.0+0x318b0)
#43 0x7ffff70be972 in g_signal_emit (/lib64/libgobject-2.0.so.0+0x31972)
#44 0x7ffff79568e1 (/lib64/libgtk-3.so.0+0x3568e1)
#45 0x7ffff7962f3d (/lib64/libgtk-3.so.0+0x362f3d)
#46 0x7ffff70ad439 in g_object_run_dispose (/lib64/libgobject-2.0.so.0+0x20439)
#47 0x3d34b7 in panel_shell_quit /home/s/code/mate-panel/mate-panel/panel-shell.c:161:3
#48 0x3d34b7 in panel_shell_on_name_lost /home/s/code/mate-panel/mate-panel/panel-shell.c:59:2
#49 0x7ffff71edb0f (/lib64/libgio-2.0.so.0+0x102b0f)
#50 0x7ffff6fa298c (/lib64/libglib-2.0.so.0+0x6298c)
#51 0x7ffff6f9c26b (/lib64/libglib-2.0.so.0+0x5c26b)
#52 0x7ffff6ffd2a7 (/lib64/libglib-2.0.so.0+0xbd2a7)
#53 0x7ffff6fa2226 in g_main_loop_run (/lib64/libglib-2.0.so.0+0x62226)
#54 0x7ffff77ee734 in gtk_main (/lib64/libgtk-3.so.0+0x1ee734)
#55 0x39f499 in main /home/s/code/mate-panel/mate-panel/main.c:223:2
#56 0x7ffff6b09087 in __libc_start_call_main (/lib64/libc.so.6+0x2a087)
#57 0x7ffff6b0914a in __libc_start_main@GLIBC_2.2.5 (/lib64/libc.so.6+0x2a14a)
#58 0x2c2884 in _start (/usr/local/bin/mate-panel+0x2c2884)
0x5160000920c0 is located 576 bytes inside of 584-byte region [0x516000091e80,0x5160000920c8)
freed by thread T0 here:
#0 0x360ada in free (/usr/local/bin/mate-panel+0x360ada)
#1 0x7ffff6f9ce54 in g_free (/lib64/libglib-2.0.so.0+0x5ce54)
#2 0x7ffff7f5aff9 in mate_image_menu_item_finalize /home/s/code/mate-desktop/libmate-desktop/mate-image-menu-item.c:353:5
#3 0x7ffff70ad283 in g_object_unref (/lib64/libgobject-2.0.so.0+0x20283)
#4 0x7ffff7800dd4 (/lib64/libgtk-3.so.0+0x200dd4)
#5 0x7ffff7708272 (/lib64/libgtk-3.so.0+0x108272)
#6 0x7ffff709e43b in g_closure_invoke (/lib64/libgobject-2.0.so.0+0x1143b)
#7 0x7ffff70cde54 (/lib64/libgobject-2.0.so.0+0x40e54)
#8 0x7ffff70be653 (/lib64/libgobject-2.0.so.0+0x31653)
#9 0x7ffff70be8b0 in g_signal_emit_valist (/lib64/libgobject-2.0.so.0+0x318b0)
#10 0x7ffff70be972 in g_signal_emit (/lib64/libgobject-2.0.so.0+0x31972)
#11 0x7ffff79568e1 (/lib64/libgtk-3.so.0+0x3568e1)
#12 0x7ffff70ad439 in g_object_run_dispose (/lib64/libgobject-2.0.so.0+0x20439)
#13 0x7ffff7788567 (/lib64/libgtk-3.so.0+0x188567)
#14 0x7ffff7708272 (/lib64/libgtk-3.so.0+0x108272)
#15 0x7ffff709e43b in g_closure_invoke (/lib64/libgobject-2.0.so.0+0x1143b)
#16 0x7ffff70cde54 (/lib64/libgobject-2.0.so.0+0x40e54)
#17 0x7ffff70be653 (/lib64/libgobject-2.0.so.0+0x31653)
#18 0x7ffff70be8b0 in g_signal_emit_valist (/lib64/libgobject-2.0.so.0+0x318b0)
#19 0x7ffff70be972 in g_signal_emit (/lib64/libgobject-2.0.so.0+0x31972)
#20 0x7ffff79568e1 (/lib64/libgtk-3.so.0+0x3568e1)
#21 0x7ffff70ad439 in g_object_run_dispose (/lib64/libgobject-2.0.so.0+0x20439)
#22 0x7ffff7708272 (/lib64/libgtk-3.so.0+0x108272)
#23 0x7ffff709e43b in g_closure_invoke (/lib64/libgobject-2.0.so.0+0x1143b)
#24 0x7ffff70cde54 (/lib64/libgobject-2.0.so.0+0x40e54)
#25 0x7ffff70be653 (/lib64/libgobject-2.0.so.0+0x31653)
#26 0x7ffff70be8b0 in g_signal_emit_valist (/lib64/libgobject-2.0.so.0+0x318b0)
#27 0x7ffff70be972 in g_signal_emit (/lib64/libgobject-2.0.so.0+0x31972)
previously allocated by thread T0 here:
#0 0x360ebc in calloc (/usr/local/bin/mate-panel+0x360ebc)
#1 0x7ffff6fa3b51 in g_malloc0 (/lib64/libglib-2.0.so.0+0x63b51)
#2 0x7ffff70c9fa5 in g_type_create_instance (/lib64/libgobject-2.0.so.0+0x3cfa5)
#3 0x7ffff70af823 (/lib64/libgobject-2.0.so.0+0x22823)
#4 0x7ffff70b0de5 in g_object_new_with_properties (/lib64/libgobject-2.0.so.0+0x23de5)
#5 0x7ffff70b1d80 in g_object_new (/lib64/libgobject-2.0.so.0+0x24d80)
#6 0x3df3fc in panel_place_menu_item_new /home/s/code/mate-panel/mate-panel/panel-menu-items.c:1445:13
#7 0x3da183 in panel_menu_bar_init /home/s/code/mate-panel/mate-panel/panel-menu-bar.c:177:31
#8 0x7ffff70ca049 in g_type_create_instance (/lib64/libgobject-2.0.so.0+0x3d049)
#9 0x7ffff70af823 (/lib64/libgobject-2.0.so.0+0x22823)
#10 0x7ffff70b0de5 in g_object_new_with_properties (/lib64/libgobject-2.0.so.0+0x23de5)
#11 0x7ffff70b1d80 in g_object_new (/lib64/libgobject-2.0.so.0+0x24d80)
#12 0x3d9421 in panel_menu_bar_load /home/s/code/mate-panel/mate-panel/panel-menu-bar.c:344:12
#13 0x3d9421 in panel_menu_bar_load_from_gsettings /home/s/code/mate-panel/mate-panel/panel-menu-bar.c:377:2
#14 0x3b7400 in mate_panel_applet_load_idle_handler /home/s/code/mate-panel/mate-panel/applet.c:946:3
#15 0x7ffff6fa298c (/lib64/libglib-2.0.so.0+0x6298c)
#16 0x7ffff6f9c26b (/lib64/libglib-2.0.so.0+0x5c26b)
#17 0x7ffff6ffd2a7 (/lib64/libglib-2.0.so.0+0xbd2a7)
#18 0x7ffff6fa2226 in g_main_loop_run (/lib64/libglib-2.0.so.0+0x62226)
#19 0x7ffff77ee734 in gtk_main (/lib64/libgtk-3.so.0+0x1ee734)
#20 0x39f499 in main /home/s/code/mate-panel/mate-panel/main.c:223:2
#21 0x7ffff6b09087 in __libc_start_call_main (/lib64/libc.so.6+0x2a087)
#22 0x7ffff6b0914a in __libc_start_main@GLIBC_2.2.5 (/lib64/libc.so.6+0x2a14a)
#23 0x2c2884 in _start (/usr/local/bin/mate-panel+0x2c2884)
SUMMARY: AddressSanitizer: heap-use-after-free /home/s/code/mate-desktop/libmate-desktop/mate-image-menu-item.c:354:29 in mate_image_menu_item_finalize Shadow bytes around the buggy address: 0x516000091e00: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa 0x516000091e80: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd 0x516000091f00: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd 0x516000091f80: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd 0x516000092000: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd =>0x516000092080: fd fd fd fd fd fd fd fd[fd]fa fa fa fa fa fa fa 0x516000092100: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa 0x516000092180: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd 0x516000092200: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd 0x516000092280: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd 0x516000092300: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd Shadow byte legend (one shadow byte represents 8 application bytes): Addressable: 00 Partially addressable: 01 02 03 04 05 06 07 Heap left redzone: fa Freed heap region: fd Stack left redzone: f1 Stack mid redzone: f2 Stack right redzone: f3 Stack after return: f5 Stack use after scope: f8 Global redzone: f9 Global init order: f6 Poisoned by user: f7 Container overflow: fc Array cookie: ac Intra object redzone: bb ASan internal: fe Left alloca redzone: ca Right alloca redzone: cb Here's the crashing code: 348 static void
349 mate_image_menu_item_finalize (GObject *object)
350 {
351 MateImageMenuItem *image_menu_item = MATE_IMAGE_MENU_ITEM (object);
352
353 g_free (image_menu_item->label);
354 image_menu_item->label = NULL;
355
356 G_OBJECT_CLASS (mate_image_menu_item_parent_class)->finalize (object);
357 }The report seems to indicate that the This leads to a use-after-free on the next line (354): Here are the addresses from GDB right before the crash: The address of the label string ( The address of I commented out the Both of these fixes make me think that the different struct layouts of |
|
Brilliant analyses. |
|
Notice that for MateImageMenuItem we have a very simple structure containing an internal gtkmenuitem, an image, and a label only
For GtkImageMenuItem we have a structure containining a GtkImageMenuItemPrivate which is defined elsewhere. It thus stands to
reason that using g_free the way that frees an internal element in GtkImageMenuItem would free the whole thing in our simpler
structure.
|
|
The issue is not with The issue is that mentioned commit only change the GType used as the parent, without adjusting the derived structures ( However, it's not that easy, because currently So… this needs to change. So the code could be updated like so: diff --git a/mate-panel/panel-menu-items.h b/mate-panel/panel-menu-items.h
index 1b94a2d8..c48d55b8 100644
--- a/mate-panel/panel-menu-items.h
+++ b/mate-panel/panel-menu-items.h
@@ -43,13 +43,13 @@ typedef struct _PanelPlaceMenuItemClass PanelPlaceMenuItemClass;
typedef struct _PanelPlaceMenuItemPrivate PanelPlaceMenuItemPrivate;
struct _PanelPlaceMenuItem {
- GtkImageMenuItem menuitem;
+ MateImageMenuItem menuitem;
PanelPlaceMenuItemPrivate *priv;
};
struct _PanelPlaceMenuItemClass {
- GtkImageMenuItemClass menuitem_class;
+ MateImageMenuItemClass menuitem_class;
};
GType panel_place_menu_item_get_type (void) G_GNUC_CONST;
@@ -66,13 +66,13 @@ typedef struct _PanelDesktopMenuItemClass PanelDesktopMenuItemClass;
typedef struct _PanelDesktopMenuItemPrivate PanelDesktopMenuItemPrivate;
struct _PanelDesktopMenuItem{
- GtkImageMenuItem menuitem;
+ MateImageMenuItem menuitem;
PanelDesktopMenuItemPrivate *priv;
};
struct _PanelDesktopMenuItemClass {
- GtkImageMenuItemClass menuitem_class;
+ MateImageMenuItemClass menuitem_class;
};
GType panel_desktop_menu_item_get_type (void) G_GNUC_CONST; |
|
Aside from that though, I cannot understand why This however doesn't change anything to the issue at hand1. Footnotes
|
|
IMO, we need to do 2 things:
|
Do we need to revert ebb3a79 too, or is this independent from used GType? I will check if we do not use Any way go ahead and don't wait for a review from me as i travel a few days. I did already confirm that reverting the commit fixes the crash. |
|
I did a PR for reverting that commit. |
@cwendling |
|
Wondering if a new minimum version of mate-desktop required AFTER the 1.28 release might be a problem. |
|
OK, the Gtype is incorrect, so that commit itself does have to go for 1.28. Problem is we then get the conflict between GtkImageMenuItem and MateImageMenuItem, which caused the "places" and "system" menuitem icons to disappear in Thus we need to do two things at once: revert 675f72f which is in fact incomplete due to the remaining use of GtkImageMenu item in the header file. That being done, we need to switch those two menu items back to GtkImageMenuItem for 1.28, or to pack the icons into a straight GtkMenuItem ourselves, basically duplicating this from MateImageMenuItem directly into the panel code to work around the library versioning issue for 1.28. This mess is necessary because we just released, but we would have the option of moving this back to MateImageMenuItem for 1.29. |
|
#1434 with my second commit to property use GtkImageMenuItem where we already refer to in in panel-menus.h needs to be tested with |
|
@lukefromdc Make |
zhuyaliang
added a commit
to mate-desktop/mate-desktop
that referenced
this issue
Mar 4, 2024
See the discussion on:mate-desktop/mate-panel#1433
|
Fixed with mate-desktop/mate-desktop#603 |
zhuyaliang
added a commit
to mate-desktop/mate-desktop
that referenced
this issue
Mar 7, 2024
See the discussion on:mate-desktop/mate-panel#1433
lukefromdc
pushed a commit
to mate-desktop/mate-desktop
that referenced
this issue
Mar 11, 2024
See the discussion on:mate-desktop/mate-panel#1433
|
Hi @raveit65 ,
An error message appears with the following content. For example: Error I can't add the panels again either because the same error occurs. I only use Tumbleweed for testing in a VM. Greetings |
|
Which mate-panel 1.28 version do you have? The fix for this is in 1.28.1 and is NOT in 1.28.0, and it
requires mate-desktop 1.28.2. Using mate-desktop 1.28.2 with mate-panel 1.28.0 was enough to
fix this in some tests, but this was by chance not by correct code.
|
|
Hello, I have the versions: mate-panel 1.28.0-1.1 |
|
OK, you have the current mate-desktop and the older mate-desktop. That should work, either you found
a case in which it does not (one of the potential reasons for mate-panel 1.28.1) or we have another issue
here.
|
|
The error messages are totally different, so you may have found a new problem.
Can you run this under gdb by any chance? Simple way to do so is
to open two terminals, run mate-panel --replace in one and close that terminal to kill the panel and leave it terminated. Then run gdb mate-panel from the other terminal. Do this after installing debugging symbols for mate-panel or after
installing with make install, either way should give a usable backtrace
|
|
Hi, here is the output from gdb mate-panel:
´´´
***@***.***:~> gdb mate-panel
GNU gdb (GDB; openSUSE Tumbleweed) 13.2
Copyright (C) 2023 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later
<http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
Type "show copying" and "show warranty" for details.
This GDB was configured as "x86_64-suse-linux".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://bugs.opensuse.org/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from mate-panel...
Reading symbols from /usr/lib/debug/usr/bin/mate-panel.debug...
(gdb) run
Starting program: /usr/bin/mate-panel
This GDB supports auto-downloading debuginfo from the following URLs:
<https://debuginfod.opensuse.org/>
Enable debuginfod for this session? (y or [n]) n
Debuginfod has been disabled.
To make this setting permanent, add 'set debuginfod enabled off' to
.gdbinit.
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib64/libthread_db.so.1".
[New Thread 0x7ffff5a006c0 (LWP 2383)]
[New Thread 0x7ffff50006c0 (LWP 2384)]
[New Thread 0x7fffefe006c0 (LWP 2385)]
[New Thread 0x7fffef4006c0 (LWP 2386)]
[New Thread 0x7fffeea006c0 (LWP 2387)]
[New Thread 0x7fffed4006c0 (LWP 2388)]
[New Thread 0x7fffeca006c0 (LWP 2389)]
[New Thread 0x7fffe3e006c0 (LWP 2392)]
[New Thread 0x7fffe34006c0 (LWP 2394)]
[New Thread 0x7fffe2a006c0 (LWP 2395)]
[Thread 0x7fffe34006c0 (LWP 2394) exited]
[Thread 0x7fffe3e006c0 (LWP 2392) exited]
[Thread 0x7fffe2a006c0 (LWP 2395) exited]
[Thread 0x7fffeca006c0 (LWP 2389) exited]
** (mate-panel:2380): WARNING **: 12:19:48.564: Error getting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
(mate-panel:2380): GLib-CRITICAL **: 12:19:48.565: g_variant_ref:
assertion 'value != NULL' failed
** (mate-panel:2380): WARNING **: 12:19:48.565:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565: Error getting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
(mate-panel:2380): GLib-CRITICAL **: 12:19:48.565: g_variant_ref:
assertion 'value != NULL' failed
** (mate-panel:2380): WARNING **: 12:19:48.565:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565: Error setting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565: Error setting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565: Error setting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565: Error setting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565: Error setting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565: Error setting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565: Error getting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
(mate-panel:2380): GLib-CRITICAL **: 12:19:48.565: g_variant_ref:
assertion 'value != NULL' failed
** (mate-panel:2380): WARNING **: 12:19:48.565:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565: Error getting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
(mate-panel:2380): GLib-CRITICAL **: 12:19:48.565: g_variant_ref:
assertion 'value != NULL' failed
** (mate-panel:2380): WARNING **: 12:19:48.565:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565: Error setting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565: Error setting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565: Error setting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565: Error setting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565: Error setting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565: Error setting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.565:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
** (mate-panel:2380): WARNING **: 12:19:48.603: Error setting property:
GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message recipient
disconnected from message bus without replying
[Thread 0x7fffed4006c0 (LWP 2388) exited]
´´´
|
|
in order to get a backtrace (which sometimes can show the exact line of code where the crash is coming from) run |
|
Hi, I waited half an hour and then interrupted the process so I could enter the command "thread apply all bt full". Here is the output: |
|
And here with the missing "debuginfos" packages mentioned above: |
|
Looks like in this case, since mate-panel --replace wasn't being run, the crash was not invoked.
What might work is running gdb mate-panel, then run --replace at the gdb prompt without first
killing off the previous panel instance. Then again, different timing might prevent the crash if\
any kind of race condition is involved, I've seen this before.
|
|
Here it is: |
|
If and only if you get the crash, that's when you type |
|
After I enter |
|
OK, I am out of ideas but this does look like a different crash. If anything else is showing dbus errors there could be an issue there, or another issue in the panel. A git bisect to find the offending commit is another way to find this, I cannot do it as I cannot duplicate your crash |
|
Hello everyone,
here is a small update for OpenSuse Tumbleweed:
If you install the "dconf-editor" package, the panel no longer crashes.
Strange, but it works. =)
See here:
https://bugzilla.suse.com/show_bug.cgi?id=1222583
Best regards
|
|
That is indeed another issue: mate-panel depending on a schema supplied by dconf-editor, fallout from the
relocatable schema support work.
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Expected behaviour
No crash
Actual behaviour
mate-panel --replacecommand crashes the panelSteps to reproduce the behaviour
With a running mate-panel do
mate-panel --replaceBoom, the panel crashes
MATE general version
1.28.1
Package version
1.28.0
Linux Distribution
fedora 40
Link to bugreport of your Distribution (requirement)
Do not file out reports against myself.
I am still wondering why this commit cause a panel crash with --replace option.
675f72f
I reverted this and previous commit on the
fix_panel--replacebranch to confirm the issue.With this branch the crash is gone.
The text was updated successfully, but these errors were encountered: