Bug 1887266

Hello Florian

Can you help to check this issue

Thanks in advance,
Mike

The message comes from GTK/GDK when the connection to the X server is lost. Shutting down the X server when the session ends isn't a responsibility of gnome-shell, so moving to GDM.

The message is harmless, but it comes from gtk:

static int•
gdk_x_io_error (Display *display)•
{•
  /* This is basically modelled after the code in XLib. We need•
   * an explicit error handler here, so we can disable our atexit()•
   * which would otherwise cause a nice segfault.•
   * We fprintf(stderr, instead of g_warning() because g_warning()•
   * could possibly be redirected to a dialog•
   */•
  if (errno == EPIPE)•
    {•
      g_message ("The application '%s' lost its connection to the display %s;\n"•
                 "most likely the X server was shut down or you killed/destroyed\n"•
                 "the application.\n",•
                 g_get_prgname (),•
                 display ? DisplayString (display) : gdk_get_display_arg_name ());•
    }•
  else•
    {•
      g_message ("%s: Fatal IO error %d (%s) on X server %s.\n",•
                 g_get_prgname (),•
                 errno, g_strerror (errno),•
                 display ? DisplayString (display) : gdk_get_display_arg_name ());•
    }•
•
  _exit (1);•
}•

There are a number of problems with this handler:

1) It tries to rely on EPIPE to mean "x server was shut down", but EPIPE isn't the only error that a "X server was shut down" event could generate. As we see above, we also get errno = 0, and errno = 11. it's possible this check was correct pre-xcb, not sure.

2) The comment says it uses fprintf, but the code doesn't. This is because of these two commits:

commit 93e203e06769ccae494fc67809b9b4ee855f5b05
Author: Owen W. Taylor <otaylor>
Date:   Tue Sep 21 00:30:46 2010 -0400

    Use g_warning() for X errors and X IO errors
    
    Currently fprintf(stderr, ...) is used for X error and X IO errors
    (unless compiled with debugging, in which case g_error() is used for
    X errors.)
    
    But if an application is redirecting log messages, we really want
    X errors to appear in that same log, so it's better to use a g_logv()
    variant.
    
    Though g_warning() isn't really appropriate for "lost connection to the
    server", G_LOG_LEVEL_INFO isn't part of the "prefixed log levels"
    so will produce output without the application name and PID.
    
    https://bugzilla.gnome.org/show_bug.cgi?id=630216

commit c70ba3a4f0043e11fffe0023685f99ec3990c644
Author: Matthias Clasen <mclasen>
Date:   Wed Nov 30 13:43:17 2016 -0500

    x11: Don't warn if the display is closed
    
    This causes a storm of warnings from all applications in the logs
    whenever the display goes away, and is not useful.

The latter of these commits makes it clear the message isn't supposed to be showing up in the log anymore.
I think Matthias forgot that g_message does show up in logs.

My suggestion is to just make the entire function be:

      g_debug ("%s: Fatal IO error %d (%s) on X server %s.\n",•
               g_get_prgname (),•
               errno, g_strerror (errno),•
               display ? DisplayString (display) : gdk_get_display_arg_name ());
      _exit (1);

g_debug doesn't show up in logs unless the user enables debugging and does get prefixed, so it seems like what we want.

moving to gtk

Hello Ben

Can we request to add this issue to the release note known issues?

Best Regards,
Mike

Hello Ben

Any plan for the bug fixing on this issue on RHEL8.4?

Thanks in advance,
Mike

Moving to POST as there's patch available.

(In reply to Tomas Popela from comment #7)
> Moving to POST as there's patch available.

Hi, Tomas

I saw the bug in post status for several month, Is there any plan to backport the patch to RHEL?

(In reply to Mike Cao from comment #8)
> (In reply to Tomas Popela from comment #7)
> > Moving to POST as there's patch available.
> 
> Hi, Tomas
> 
> I saw the bug in post status for several month, Is there any plan to
> backport the patch to RHEL?

We are waiting for more gtk3 related fixes (we have few queued in RHEL 7 where one is waiting on upstream fix) to be brought into RHEL 8. It's possible that this will be fixed in 8.6.0 - will sync with Matthias after he's back from PTO.

Backport builds fine locall.y

Reproduced with gtk3-3.22.30-8.el8.x86_64
Cannot reproduce with gtk3-3.22.30-9.el8.x86_64.

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 (gtk3 bug fix and enhancement update), 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-2022:1946