After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 787633 - gvfsd-trash crashed with signal 5 in g_main_context_new()
gvfsd-trash crashed with signal 5 in g_main_context_new()
Status: RESOLVED INCOMPLETE
Product: gvfs
Classification: Core
Component: daemon
1.32.x
Other Linux
: Normal critical
: ---
Assigned To: gvfs-maint
gvfs-maint
: 787636 (view as bug list)
Depends on:
Blocks:
 
 
Reported: 2017-09-13 15:22 UTC by Cristian Aravena Romero
Modified: 2017-12-19 14:13 UTC
See Also:
GNOME target: ---
GNOME version: ---


Description Cristian Aravena Romero 2017-09-13 15:22:18 UTC 
Hello,

Open bug in launchpad.net
https://bugs.launchpad.net/ubuntu/+source/gvfs/+bug/1716940

"I do not know how this happened."


BackTrace:

+ Trace 237955

  • #0 _g_log_abort
    at ../../../../glib/gmessages.c line 554
  • #1 g_log_default_handler
    at ../../../../glib/gmessages.c line 3051
  • #2 g_logv
    at ../../../../glib/gmessages.c line 1341
  • #3 g_log
    at ../../../../glib/gmessages.c line 1403
  • #4 g_wakeup_new
    at ../../../../glib/gwakeup.c line 161
  • #5 g_main_context_new
    at ../../../../glib/gmain.c line 658
  • #6 initable_init
    at ../../../../gio/gdbusproxy.c line 1938
  • #7 g_initable_new_valist
    at ../../../../gio/ginitable.c line 248
  • #8 g_initable_new
    at ../../../../gio/ginitable.c line 162
  • #9 gvfs_dbus_enumerator_proxy_new_sync
  • #10 create_enumerator_proxy
    at gvfsjobenumerate.c line 125
  • #11 send_infos
    at gvfsjobenumerate.c line 158
  • #12 g_vfs_job_enumerate_add_info
    at gvfsjobenumerate.c line 209
  • #13 trash_backend_enumerate_root
    at gvfsbackendtrash.c line 612
  • #14 trash_backend_enumerate
    at gvfsbackendtrash.c line 690
  • #15 g_vfs_job_try
    at gvfsjob.c line 215
  • #16 g_vfs_daemon_queue_job
    at gvfsdaemon.c line 600
  • #17 g_closure_invoke
    at ../../../../gobject/gclosure.c line 804
  • #18 signal_emit_unlocked_R
    at ../../../../gobject/gsignal.c line 3635
  • #19 g_signal_emit_valist
    at ../../../../gobject/gsignal.c line 3391
  • #20 g_signal_emit
    at ../../../../gobject/gsignal.c line 3447
  • #21 g_vfs_job_source_new_job
    at gvfsjobsource.c line 101
  • #22 g_vfs_job_enumerate_new_handle
    at gvfsjobenumerate.c line 109
  • #23 ffi_call_unix64
    at ../src/x86/unix64.S line 76
  • #24 ffi_call
    at ../src/x86/ffi64.c line 525
  • #25 g_cclosure_marshal_generic
    at ../../../../gobject/gclosure.c line 1490
  • #26 g_closure_invoke
    at ../../../../gobject/gclosure.c line 804
  • #27 signal_emit_unlocked_R
    at ../../../../gobject/gsignal.c line 3635
  • #28 g_signal_emitv
    at ../../../../gobject/gsignal.c line 3129
  • #29 _gvfs_dbus_mount_skeleton_handle_method_call
    at gvfsdbus.c line 15259
  • #30 g_dbus_interface_method_dispatch_helper
    at ../../../../gio/gdbusinterfaceskeleton.c line 609
  • #31 call_in_idle_cb
    at ../../../../gio/gdbusconnection.c line 4851
  • #32 g_main_dispatch
    at ../../../../glib/gmain.c line 3148
  • #33 g_main_context_dispatch
    at ../../../../glib/gmain.c line 3813
  • #34 g_main_context_iterate
    at ../../../../glib/gmain.c line 3886
  • #35 g_main_loop_run
    at ../../../../glib/gmain.c line 4082
  • #36 daemon_main
    at daemon-main.c line 398
  • #37 main
    at daemon-main-generic.c line 45 

Regards,
--
Cristian
Comment 1 Ondrej Holy 2017-09-14 07:34:32 UTC 
*** Bug 787636 has been marked as a duplicate of this bug. ***
Comment 2 Ondrej Holy 2017-09-14 09:13:34 UTC 
It fails with "Creating pipes for GWakeup: Too many open files". It might mean that we are leaking file descriptors somewhere, but I don't see anything unusual using "lsof -c gvfs". You reported two bugs for different processes at the same time, so I suppose that system limit was reached, so it can be caused by whatever process...

Please let me know if you see some enormous file descriptor usage by gvfs processes:
printf "  count\tpid\tname\n"; lsof -c gvfs | awk '{ if (NR > 1) print $2 "\t" $1; }' | sort | uniq -c | sort -rn

But I suppose it was caused by something else, you can show the top 10 file descriptor consumers: 
printf "  count\tpid\tname\n"; lsof | awk '{ if (NR > 1) print $2 "\t" $1; }' | sort | uniq -c | sort -rn | head -10

(I hope the commands are ok, I wrote them in hurry.)
Comment 3 Ondrej Holy 2017-09-14 09:18:27 UTC 
Just a note that you can check the limits using the following:

- per process limit:
ulimit -n

- system limit:
cat /proc/sys/fs/file-max
Comment 4 Cristian Aravena Romero 2017-09-14 15:47:09 UTC 
Hello,

$ ulimit -n
1024

$ cat /proc/sys/fs/file-max
1198066

Regards,
--
Cristian
Comment 5 Ondrej Holy 2017-09-16 13:12:14 UTC 
Thanks, those limits looks ok. Do you see some enormous file descriptor usage by gvfs processes (Comment #2)? Do you see those crashes regularly?
Comment 6 André Klapper 2017-12-19 14:13:42 UTC 
Closing this bug report as no further information has been provided. Please feel free to reopen this bug report if you can provide the information that was asked for in a previous comment.
Thanks!