Red Hat Bugzilla – Bug 1489359
SIGSEGV in GDM when SELinux PAM setup fails (Unable to get valid context for USERNAME)
Last modified: 2018-05-17 10:39:45 EDT
Description of problem: GDM crashes on some login failure scenarios. This makes it harder to debug the real issue. I've opened this issue so you can consider checking this pointer. Version-Release number of selected component (if applicable): gdm-3.22.3-12.el7.x86_64 How reproducible: Currently unknown Steps to Reproduce: 1. Upgrade from 7.3 to 7.4, and get an SELinux issue :) pam_selinux(gdm-password:session): Unable to get valid context for USERNAME 2. Try to log in 3. Actual results: GDM crashes in remove_user_session because manager->priv->user_sessions is null. Expected results: GDM should not crash when PAM setup fails Additional info: Core was generated by `/usr/sbin/gdm'. Program terminated with signal 11, Segmentation fault. #0 0x00005622e431e872 in remove_user_session (manager=0x5622e5361e00, session=0x5622e5348100) at gdm-manager.c:1959 1959 node = g_list_find (manager->priv->user_sessions, session); (gdb) print manager->priv->user_sessions $2 = (GList *) 0x0 #0 0x00005622e431e872 in remove_user_session (manager=0x5622e5361e00, session=0x5622e5348100) at gdm-manager.c:1959 #1 0x00007f720de6edcc in ffi_call_unix64 () from /lib64/libffi.so.6 #2 0x00007f720de6e6f5 in ffi_call () from /lib64/libffi.so.6 #3 0x00007f720f1ec1a8 in g_cclosure_marshal_generic () from /lib64/libgobject-2.0.so.0 #4 0x00007f720f1eb968 in g_closure_invoke () from /lib64/libgobject-2.0.so.0 #5 0x00007f720f1fda7d in signal_emit_unlocked_R () from /lib64/libgobject-2.0.so.0 #6 0x00007f720f205af1 in g_signal_emit_valist () from /lib64/libgobject-2.0.so.0 #7 0x00007f720f205ddf in g_signal_emit () from /lib64/libgobject-2.0.so.0 #8 0x00005622e432766b in on_opened (worker=<optimized out>, res=<optimized out>, user_data=0x5622e535d9c0) at gdm-session.c:918 #9 0x00007f720f4b6f64 in g_task_return_now () from /lib64/libgio-2.0.so.0 #10 0x00007f720f4b760e in g_task_return () from /lib64/libgio-2.0.so.0 #11 0x00007f720f4f5eb9 in reply_cb () from /lib64/libgio-2.0.so.0 #12 0x00007f720f4b6f64 in g_task_return_now () from /lib64/libgio-2.0.so.0 #13 0x00007f720f4b760e in g_task_return () from /lib64/libgio-2.0.so.0 #14 0x00007f720f4ea9d9 in g_dbus_connection_call_done () from /lib64/libgio-2.0.so.0 #15 0x00007f720f4b6f64 in g_task_return_now () from /lib64/libgio-2.0.so.0 #16 0x00007f720f4b6f99 in complete_in_idle_cb () from /lib64/libgio-2.0.so.0 #17 0x00007f720ef110d7 in g_idle_dispatch () from /lib64/libglib-2.0.so.0 #18 0x00007f720ef144c9 in g_main_context_dispatch () from /lib64/libglib-2.0.so.0 #19 0x00007f720ef14818 in g_main_context_iterate.isra.21 () from /lib64/libglib-2.0.so.0 #20 0x00007f720ef14aea in g_main_loop_run () from /lib64/libglib-2.0.so.0 #21 0x00005622e4314914 in main (argc=1, argv=0x7ffed8fbf678) at main.c:404 journal: gnome-shell: Fatal IO error 11 (Resource temporarily unavailable) on X server :0. journal: Name taken or bus went away - shutting down gnome-session: gnome-session-binary[2167]: WARNING: Lost name on bus: org.gnome.SessionManager gnome-session-binary[2167]: WARNING: Lost name on bus: org.gnome.SessionManager journal: gnome-settings-daemon: Fatal IO error 2 (No such file or directory) on X server :0. gdm: Child process -2167 was already dead. kernel: traps: gdm[1445] general protection ip:55b7d246d872 sp:7ffd19f6de20 error:0 in gdm[55b7d2454000+5e000] abrt-hook-ccpp: Process 1445 (gdm) of user 0 killed by SIGSEGV - dumping core systemd: gdm.service: main process exited, code=dumped, status=11/SEGV systemd: Unit gdm.service entered failed state. systemd: Triggering OnFailure= dependencies of gdm.service. systemd: gdm.service failed.
NULL is a valid GList, so there must be more to the story here. Do you mind attaching the core file?
Sent directly
Thanks, it looks like the problem is fixed with this commit: https://git.gnome.org/browse/gdm/commit/?id=0ab4f63
I have this problem. How can I fix it? Now I can only login with root account. Creating a new account also doesn't work.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2018:0770