Description of problem: after re installing fedora I faced this problem. I was thinking its HD or ram issue Version-Release number of selected component: systemd-204-17.fc19 Additional info: reporter: libreport-2.1.9 backtrace_rating: 4 cmdline: /usr/lib/systemd/systemd-journald crash_function: journal_file_find_data_object_with_hash executable: /usr/lib/systemd/systemd-journald kernel: 3.11.9-200.fc19.x86_64 runlevel: N 5 type: CCpp uid: 0 Truncated backtrace: Thread no. 1 (10 frames) #0 journal_file_find_data_object_with_hash at ../src/journal/journal-file.c:803 #1 journal_file_append_data at ../src/journal/journal-file.c:958 #2 journal_file_append_entry at ../src/journal/journal-file.c:1336 #3 write_to_journal at ../src/journal/journald-server.c:476 #4 dispatch_message_real at ../src/journal/journald-server.c:682 #5 server_dispatch_message at ../src/journal/journald-server.c:776 #6 stdout_stream_log at ../src/journal/journald-stream.c:130 #7 stdout_stream_line at ../src/journal/journald-stream.c:231 #8 stdout_stream_scan at ../src/journal/journald-stream.c:261 #9 stdout_stream_process at ../src/journal/journald-stream.c:312
Created attachment 840271 [details] File: backtrace
Created attachment 840272 [details] File: cgroup
Created attachment 840273 [details] File: core_backtrace
Created attachment 840274 [details] File: dso_list
Created attachment 840275 [details] File: environ
Created attachment 840276 [details] File: exploitable
Created attachment 840277 [details] File: limits
Created attachment 840278 [details] File: maps
Created attachment 840279 [details] File: open_fds
Created attachment 840280 [details] File: proc_pid_status
Created attachment 840281 [details] File: var_log_messages
Sorry for the slow response. Yes, mmaped files cause segfaults on IO. Not much we can do about this without kernel changes.
*** Bug 1177726 has been marked as a duplicate of this bug. ***
*** Bug 1164427 has been marked as a duplicate of this bug. ***
*** Bug 1151848 has been marked as a duplicate of this bug. ***
*** Bug 1098722 has been marked as a duplicate of this bug. ***
I now added some code to journald git, that catches SIGBUS, remaps the triggering page to an anonymous page, then detects this afterwards and rotates the file away. This should make us more robust on file systems where overwriting bytes in a file can fail due to disk full, such as COW file systems (btrfs).
*** Bug 1053292 has been marked as a duplicate of this bug. ***
http://cgit.freedesktop.org/systemd/systemd/commit/?id=fa6ac76083b8ffc1309876459f54f9f0e2843731 This patch is nontrivial to backport, even to F20.