Description of problem: I was running xfs_irecover. The device file was read-only ("br--r--r--."). Version-Release number of selected component: xfsprogs-3.2.1-1.fc20 Additional info: reporter: libreport-2.2.3 backtrace_rating: 4 cmdline: xfs_db -r /dev/disk/by-id/ata-ST31000340AS_9QJ03EZM-part1 crash_function: getbit executable: /usr/sbin/xfs_db kernel: 3.16.3-200.fc20.x86_64 runlevel: N 5 type: CCpp uid: 5000 Truncated backtrace: Thread no. 1 (10 frames) #0 getbit at bit.c:39 #1 getbitval at bit.c:109 #2 fp_num at fprint.c:88 #3 print_sarray at print.c:206 #4 fp_sarray at fprint.c:122 #5 print_flist_1 at print.c:143 #8 print_flist at print.c:95 #9 print_allfields at print.c:62 #10 print_struct at print.c:292 #11 print_f at print.c:84
Created attachment 940237 [details] File: backtrace
Created attachment 940238 [details] File: cgroup
Created attachment 940239 [details] File: core_backtrace
Created attachment 940240 [details] File: dso_list
Created attachment 940241 [details] File: environ
Created attachment 940242 [details] File: exploitable
Created attachment 940243 [details] File: limits
Created attachment 940244 [details] File: maps
Created attachment 940245 [details] File: open_fds
Created attachment 940246 [details] File: proc_pid_status
Created attachment 940247 [details] File: var_log_messages
This one may be tough to sort out w/o a reproducer. Do you still have the filesystem that caused it?
Unfortunately I don't have the filesystem anymore and I also didn't have the inspiration to run xfs_metadump on it. Though I'm not sure how much it would have helped since all the files were supposedly removed by rm, if I remember correctly.
Ok, and this was xfs_db as driven by xfs_irecover, which is groveling around the disk looking at things that may not even be intact inodes. So there's some error handling or disk structure verification missing somewhere ... but I'm not sure how to work backwards to what that was. Going to have to close this one, sorry.