Good evening,
I'm running PVFS with two I/O nodes and a single metadata node. I
think I've come across some wierdness in the values returned by
statfs. My I/O nodes are mounting directories on RAID arrays that are
3.2TB (2.8TB free) and 2.0TB (1.4TB) free. For the PVFS client mount
point, df displays something sane for this configuration:
$ df
Filesystem Size Used Avail Use% Mounted on
/dev/hda1 9.7G 6.9G 2.3G 76% /
/dev/hda6 45G 668M 44G 2% /home
/dev/sda 3.2T 461G 2.8T 15% /RAIDS/RAID_1
tcp://cluster-1:3334/pvfs2-fs
3.9T 1.2T 2.7T 30% /RAIDS/RAID_2
However, many apps, like Nautilus, are report that the mount point has
far less free space -- 688MB. It reports this no matter how many
files we add to the PVFS2 filesystem.
If I make a statfs function call, I get the smaller amounts. To wit:
$ python
Python 2.4 (#2, Feb 12 2005, 00:29:46)
[GCC 3.4.3 (Mandrakelinux 10.2 3.4.3-3mdk)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import os
>>> os.statvfs('/RAIDS/RAID_2').f_blocks
1001258L
>>> os.statvfs('/RAIDS/RAID_2').f_bfree
705068L
(using strace, the system call that Python's os.statvfs actually makes
is:
statfs64("/RAIDS/RAID_2", 84, {f_type=0x20030528, f_bsize=4194304,
f_blocks=1001258, f_bfree=705068, f_bavail=705068, f_files=4294967292,
f_ffree=4294967277, f_fsid={2016975803, 0}, f_namelen=255,
f_frsize=1024}) = 0
)
705,068 1KB blocks would give me the 688MB free.
We noticed this problem when we ran one of our Windows applications
that is accessing PVFS2 via Samba. It allocates a big sparse file
then fills it. Whenever it reached the 688MB mark, it would die and
complain that there's no space left.
Any ideas what's up? Thanks,
--Justin
_______________________________________________
PVFS2-users mailing list
[email protected]
http://www.beowulf-underground.org/mailman/listinfo/pvfs2-users
