I'm looking into a problem in one of the openQA tests - the desktop_login test - and it seems related to this crash.
That test tests creating a couple of different user accounts and logging in, logging out, user switching and so on. Sometimes, in F36 and Rawhide, when we try to log out of an account, it takes way longer than expected to actually end the session and go back to GDM - about a minute and a half (usually it's a few seconds).
I grabbed the logs from one such failure and found what looks like this same crash logged right around the time the session finally ended and we returned to GDM:
Thread 1 (Thread 0x7ff8e3f475c0 (LWP 3631)):
#0 meta_context_terminate (context=0x0) at ../src/core/meta-context.c:451 _g_boolean_var_ = <optimized out>
priv = 0xffffffffffffffa0
__func__ = "meta_context_terminate"
#1 0x00007ff8ea67ba86 in process_ice_messages (channel=<optimized out>, condition=<optimized out>, user_data=0x7ff8ed0de6f0) at ../src/x11/session.c:116 ice_connection = <optimized out>
connection = <optimized out>
status = <optimized out>
#2 process_ice_messages (channel=<optimized out>, condition=<optimized out>, user_data=0x7ff8ed0de6f0) at ../src/x11/session.c:95 ice_connection = 0x7ff8ed0de6f0
connection = <optimized out>
status = <optimized out>
#3 0x00007ff8eb2ca1bf in g_main_dispatch (context=0x7ff8ecc614f0) at ../glib/gmain.c:3413
dispatch = 0x7ff8eb31d950 <g_io_unix_dispatch>
prev_source = 0x0 begin_time_nsec = 543677000869
was_in_call = 0
user_data = 0x7ff8ed0de6f0
callback = 0x7ff8ea67ba30 <process_ice_messages>
cb_funcs = 0x7ff8eb3af3e0 <g_source_callback_funcs>
cb_data = 0x7ff8ed512480 need_destroy = <optimized out>
source = 0x7ff8ecf7b0e0
current = 0x7ff8ecc9fe60
i = 0
__func__ = <optimized out>
#4 g_main_context_dispatch (context=0x7ff8ecc614f0) at ../glib/gmain.c:4131
No locals.
#5 0x00007ff8eb31f2d8 in g_main_context_iterate.constprop.0 (context=0x7ff8ecc614f0, block=block@entry=1, dispatch=dispatch@entry=1, self=<optimized out>) at ../glib/gmain.c:4207 max_priority = 2147483647
timeout = 102634
some_ready = 1
nfds = 13 allocated_nfds = <optimized out>
fds = <optimized out> begin_time_nsec = 489151985844
#6 0x00007ff8eb2c98ef in g_main_loop_run (loop=0x7ff8ee8c5b50) at ../glib/gmain.c:4405
self = <optimized out>
__func__ = "g_main_loop_run"
#7 0x00007ff8ea642289 in meta_context_run_main_loop (context=<optimized out>, error=0x7ffd82a13550) at ../src/core/meta-context.c:433
priv = 0x7ff8ecc5c000
__func__ = "meta_context_run_main_loop"
#8 0x00007ff8eb8b8dd1 in main (argc=<optimized out>, argv=<optimized out>) at ../src/main.c:563
context = 0x7ff8ecc5c060
error = 0x0
ecode = 0
I note there's also an Ubuntu bug for a similar traceback:
I'm looking into a problem in one of the openQA tests - the desktop_login test - and it seems related to this crash.
That test tests creating a couple of different user accounts and logging in, logging out, user switching and so on. Sometimes, in F36 and Rawhide, when we try to log out of an account, it takes way longer than expected to actually end the session and go back to GDM - about a minute and a half (usually it's a few seconds).
I grabbed the logs from one such failure and found what looks like this same crash logged right around the time the session finally ended and we returned to GDM:
Thread 1 (Thread 0x7ff8e3f475c0 (LWP 3631)): terminate (context=0x0) at ../src/ core/meta- context. c:451
_g_boolean_ var_ = <optimized out> terminate" ice_messages (channel=<optimized out>, condition= <optimized out>, user_data= 0x7ff8ed0de6f0) at ../src/ x11/session. c:116
ice_connection = <optimized out> ice_messages (channel=<optimized out>, condition= <optimized out>, user_data= 0x7ff8ed0de6f0) at ../src/ x11/session. c:95
ice_connection = 0x7ff8ed0de6f0 0x7ff8ecc614f0) at ../glib/ gmain.c: 3413 dispatch>
begin_ time_nsec = 543677000869 ice_messages> callback_ funcs>
need_destroy = <optimized out> context_ dispatch (context= 0x7ff8ecc614f0) at ../glib/ gmain.c: 4131 context_ iterate. constprop. 0 (context= 0x7ff8ecc614f0, block=block@ entry=1, dispatch= dispatch@ entry=1, self=<optimized out>) at ../glib/ gmain.c: 4207
max_priority = 2147483647
allocated_ nfds = <optimized out>
begin_ time_nsec = 489151985844 c5b50) at ../glib/ gmain.c: 4405 run_main_ loop (context=<optimized out>, error=0x7ffd82a 13550) at ../src/ core/meta- context. c:433 run_main_ loop"
#0 meta_context_
priv = 0xffffffffffffffa0
__func__ = "meta_context_
#1 0x00007ff8ea67ba86 in process_
connection = <optimized out>
status = <optimized out>
#2 process_
connection = <optimized out>
status = <optimized out>
#3 0x00007ff8eb2ca1bf in g_main_dispatch (context=
dispatch = 0x7ff8eb31d950 <g_io_unix_
prev_source = 0x0
was_in_call = 0
user_data = 0x7ff8ed0de6f0
callback = 0x7ff8ea67ba30 <process_
cb_funcs = 0x7ff8eb3af3e0 <g_source_
cb_data = 0x7ff8ed512480
source = 0x7ff8ecf7b0e0
current = 0x7ff8ecc9fe60
i = 0
__func__ = <optimized out>
#4 g_main_
No locals.
#5 0x00007ff8eb31f2d8 in g_main_
timeout = 102634
some_ready = 1
nfds = 13
fds = <optimized out>
#6 0x00007ff8eb2c98ef in g_main_loop_run (loop=0x7ff8ee8
self = <optimized out>
__func__ = "g_main_loop_run"
#7 0x00007ff8ea642289 in meta_context_
priv = 0x7ff8ecc5c000
__func__ = "meta_context_
#8 0x00007ff8eb8b8dd1 in main (argc=<optimized out>, argv=<optimized out>) at ../src/main.c:563
context = 0x7ff8ecc5c060
error = 0x0
ecode = 0
I note there's also an Ubuntu bug for a similar traceback:
https:/ /bugs.launchpad .net/ubuntu/ +source/ gnome-shell/ +bug/1959937