Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: https://bugzilla.lustre.org/show_bug.cgi?id=11414 Created an attachment (id=9230) Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: --> (https://bugzilla.lustre.org/attachment.cgi?id=9230&action=view) do better consistency checking of extents headers This patch adds better consistency checking to the extents headers in all of the called extents routines. It also verifies that the depth of the extent index is decreasing as it walks the extent tree. I''ve done preliminary testing using some synthetic test cases, but not any testing on heavily corrupted filesystems. It should be used in place of the "e2fsprogs-extents.patch" in the e2fsprogs series file, or you can use the e2fsprogs-1.39-cfs2-1 RPMs available via ftp.lustre.org/pub/lustre/other/e2fsprogs/testing (currently being built & uploaded).
Please don''t reply to lustre-devel. Instead, comment in Bugzilla by
using the following link:
https://bugzilla.lustre.org/show_bug.cgi?id=11414
Event on pigs40 sdb.
(20:20:50) adilger: heidelberg1: so in the current bad inode case, please do
''debugfs -c /dev/sdb -R "imap <84951045>"'' to
get the block number, "dd
if=/dev/sdb of=/tmp/ino84951045.bin bs=4k skip={blocknr} count=1" to dump
the
inode, and then attach /tmp/ino84951045.bin to bug 11414
debugfs -c /dev/sdb -R "imap <84951045>"
debugfs 1.39.cfs2 (14-Nov-2006)
/dev/sdb: catastrophic mode - not reading inode or group bitmaps
Inode 84951045 is part of block group 20740
located at block 679608322, offset 0x0400
# pigs40 /root > dd if=/dev/sdb of=/tmp/ino84951045.bin bs=4k skip=679608322
count=1
1+0 records in
1+0 records out
I''ll attach ino84951045.bin next.
Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: https://bugzilla.lustre.org/show_bug.cgi?id=11414 Created an attachment (id=9240) Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: --> (https://bugzilla.lustre.org/attachment.cgi?id=9240&action=view) dd of offending block
Please don''t reply to lustre-devel. Instead, comment in Bugzilla by
using the following link:
https://bugzilla.lustre.org/show_bug.cgi?id=11414
# pigs40 /tmp > debugfs -w /dev/sdb
debugfs 1.39.cfs2 (14-Nov-2006)
debugfs: mi <84951045>
Mode [0106666]
User ID [7115]
Group ID [7115]
Size [0]
Creation time [1166582798]
Modification time [1166268537]
Access time [1166268537]
Deletion time [0]
Link count [1]
Block count high [0]
Block count [0]
File flags [0x80000] 0
Generation [0x427dba0]
File acl [0]
High 32bits of size [0]
Fragment address [0]
Direct Block #0 [127754] 0
Direct Block #1 [65540] 0
Direct Block #2 [1] 0
Direct Block #3 [0]
Direct Block #4 [0]
Direct Block #5 [673775616] 0
Direct Block #6 [0]
Direct Block #7 [0]
Direct Block #8 [0]
Direct Block #9 [0]
Direct Block #10 [0]
Direct Block #11 [0]
Indirect Block [0]
Double Indirect Block [0]
Triple Indirect Block [0]
debugfs: quit