Description of problem: It suddenly disappeared Version-Release number of selected component: libreoffice-core-4.2.4.2-8.fc20 Additional info: reporter: libreport-2.2.2 backtrace_rating: 4 cmdline: /usr/lib64/libreoffice/program/soffice.bin jrc2014_course_day3_5_postgis_qgis_wms.odp --splash-pipe=5 crash_function: magazine_chain_pop_head executable: /usr/lib64/libreoffice/program/soffice.bin kernel: 3.14.3-200.fc20.x86_64 runlevel: N 5 type: CCpp uid: 1338 Truncated backtrace: Thread no. 1 (10 frames) #0 magazine_chain_pop_head at gslice.c:545 #1 thread_memory_magazine1_alloc at gslice.c:848 #2 g_slice_alloc at gslice.c:1007 #3 g_slice_alloc0 at gslice.c:1042 #4 g_type_create_instance at gtype.c:1850 #5 g_object_new_internal at gobject.c:1746 #8 _g_local_file_new at glocalfile.c:304 #9 g_daemon_vfs_get_file_for_path at gdaemonvfs.c:394 #10 ih_event_callback at inotify-helper.c:181 #11 ip_event_dispatch at inotify-path.c:493
Created attachment 898803 [details] File: backtrace
Created attachment 898804 [details] File: cgroup
Created attachment 898805 [details] File: core_backtrace
Created attachment 898806 [details] File: dso_list
Created attachment 898807 [details] File: environ
Created attachment 898808 [details] File: exploitable
Created attachment 898809 [details] File: limits
Created attachment 898810 [details] File: maps
Created attachment 898811 [details] File: open_fds
Created attachment 898812 [details] File: proc_pid_status
Created attachment 898813 [details] File: var_log_messages
Why do you think it is gvfs bug? Do you have some reproducer? GVfs just simply calls GIO...
Still trying to clear the needinfo on this
magazine_chain_pop_head() crashing means that there's memory corruption earlier in the stack. It's probably not in the gvfs or GIO code, as we're not seeing hundreds of those crashes. It's a bug in LibreOffice.
No sure I buy that logic. But lets just go with "no one can reproduce it and the bt doesn't help fit it" rather than pointlessly batting it around.