[Ocfs2-users] Reproducing fragmentation and out of space error

[Hakan Koseoglu]

On 31 May 2012 14:25, Nick Anderson <nick at cmdln.org> wrote:
> Hakan, your script just ran me out of inodes, maybe you already had a
> fragmented file-system to start from and it triggered it faster.
Probably we had a smaller file system compared to yours. I can't
remember anymore unfortunately but probably the original file system
was less than a couple of hundreds of GB, if that, more likely tens of
GB (in 2009 it would have been 146GB SCSI or SAS disks with RAID1 or
RAID10). In the end we worked around it by putting various directories
with very large number of small files into separate mount points. This
way even if we hit the issue, it wouldn't be the end of the world.  We
have followed that route with other ocfs2 deployments and haven't had
a problem with anyone yet.