I noticed that the these two commands to list files below 5 GiB produce different results:
find . -type f -size -5368709120c
find . -type f -size -5G
Specifically the one that uses kilobyte unit (5368709120c
) returns additional files that are larger than the maximum file size returned by the one that uses the GiB unit (5G
).
From the find
manual page I read the following:
-size n[cwbkMG]
File uses n units of space. The following suffixes can be used:
`b' for 512-byte blocks (this is the default if no suffix is used)
`c' for bytes
`w' for two-byte words
`k' for Kilobytes (units of 1024 bytes)
`M' for Megabytes (units of 1048576 bytes)
`G' for Gigabytes (units of 1073741824 bytes)
The size does not count indirect blocks, but it does count blocks
in sparse files that are not actually allocated. Bear in mind that the `%k'
and `%b' format specifiers of -printf handle sparse files differently. The
`b' suffix always denotes 512-byte blocks and never 1 Kilobyte blocks,
which is different to the behaviour of -ls.
So, given that the unit of G
is 1073741824, 5G
should be 5368709120c
. Is the issue due to how sparse or indirect blocks are counted?
Thanks in advance for the help.
** UPDATE **
Found something else odd. The threshold at which the files returned are different is exactly 4 GiB:
Largest file found by -5G
:
4285018593 bytes = 3.990734548 GiB
Largest not file found by -5G
:
4299230968 bytes = 4.003970854 GiB
Files are stored on a XFS partition:
meta-data=/dev/mapper/vg_XXXXX_lv isize=256 agcount=197, agsize=268435440 blks
= sectsz=4096 attr=2, projid32bit=1
= crc=0 finobt=0
data = bsize=4096 blocks=52739701760, imaxpct=1
= sunit=16 swidth=256 blks
naming =version 2 bsize=4096 ascii-ci=0 ftype=0
log =internal bsize=4096 blocks=521728, version=2
= sectsz=4096 sunit=1 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
No comments:
Post a Comment