http://oss.sgi.com/bugzilla/show_bug.cgi?id=877
Summary: xfs_db segfault on bmap command
Product: XFS
Version: unspecified
Platform: PC
OS/Version: Linux
Status: NEW
Severity: major
Priority: P5
Component: xfsprogs
AssignedTo: xfs-masters@xxxxxxxxxxx
ReportedBy: xfs@xxxxxxxxxxx
Estimated Hours: 0.0
Classification: Unclassified
Roman <xfs@xxxxxxxxxxx> changed:
What |Removed |Added
----------------------------------------------------------------------------
Flag| |NeedInfo?
Created an attachment (id=290)
--> (http://oss.sgi.com/bugzilla/attachment.cgi?id=290)
gdb backtrace, core, executable
./xfs_db -c "inode 35932" -c bmap -r /dev/nbd0
can't seek in filesystem at bb 192014418689595392
Segmentation fault (core dumped)
backtrace:
#0 0x0804c822 in bmap (offset=0, len=0, whichfork=0, nexp=0xbfd59d3c,
bep=0xbfd59cf4) at bmap.c:102
102 if (be16_to_cpu(block->bb_level) == 0)
#0 0x0804c822 in bmap (offset=0, len=0, whichfork=0, nexp=0xbfd59d3c,
bep=0xbfd59cf4) at bmap.c:102
block = 0x0
bno = 26384997711609856
curoffset = 0
dip = 0x9d69c00
eoffset = 18446744073709551615
ep = 0x80680d0
fmt = XFS_DINODE_FMT_BTREE
fsize = 56
kp = 0x9d6b004
n = 0
nex = 1
nextbno = 0
nextents = 12793917
pp = 0x9d6b7f4
rblock = 0x9d69c64
typ = TYP_BMAPBTD
xp = 0xd17ff4
#1 0x0804cf75 in bmap_f (argc=1, argv=0x9d4fe40) at bmap.c:211
afork = 0
be = {startoff = 68719476752, startblock = 16, blockcount =
708470592249102324, flag = 13734816}
c = -1
co = 0
cosave = 0
dfork = 1
dip = 0x9d69c00
eo = 18446744073709551615
len = 54967668982521160
nex = 0
p = 0x9d4fe38 "("
whichfork = 0
#2 0x0805da3a in command (argc=1, argv=0x9d4fe40) at command.c:97
cmd = 0xbfd5b46c "bmap"
ct = 0x9d653a4
#3 0x080655b1 in main (argc=7, argv=0xbfd59e84) at init.c:168
c = 1
i = 1
done = 0
input = 0xd18304 "\300\214", <incomplete sequence \321>
v = 0x9d4fe40
Background info.
I used 1Tb drive before & setup nightly defragmentation, all worked fine. But
when I moved to 2Tb drive (mkfs+xfs_backup+xfs_restore) I found that ~35Gb
files are corrupted by xfs_fsr. Inode 35932 ("frag" says "inode 35932 actual
55085 ideal 1") will be corrupted if I try to defragment it, I reproduced it
twice. "bmap" command segfaults because of... seems broken metadata there I
didn't manage to understand precisely why. File is readed fine &
xfs_check/xfs_repair doesn't complain.
Metadata dump (~115Mb compressed) I'll made available privately upon request.
--
Configure bugmail: http://oss.sgi.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
|