About this list Date view Thread view Subject view Author view Attachment view

From: Jörn Engel (joern_at_wohnheim.fh-wedel.de)
Date: Wed 04 Aug 2004 - 11:22:47 BST

On Tue, 3 August 2004 12:28:21 -0400, Matt Ayres wrote:
> There is vunify which is part of util-vserver package. What is better for general usage is Sam's unify-dirs script. It is located at http://mirrors.paul.sladen.org/sam.vilain.net/vserver/unify-dirs.
> Just use this without the -l or -i options and it will just do hard links, which is what you want for the current implementation of cow links.

Darn! Doesn't work for me yet.

One personal problem is a slow (USB1) 300GB hard-drive that contains
some identical files. I was thinking about hashing only the first 4k
or so of each file and the do a direct comparison in case of hash
collision. Even with sha1 over the complete file, there is no
guarantee that a hash collision means two identical files.

Also, I want a database with all already known files. Untimately this
could be turned into a daemon that watches the complete fs tree for
changes and turns new files into cowlinks shortly after creation.
With such a daemon, "cp -r" will temporarily flush part of the page
cache, have the same result as "cowcopy -r".

Still interesting reading, thanks.


Data expands to fill the space available for storage.
-- Parkinson's Law
Vserver mailing list

About this list Date view Thread view Subject view Author view Attachment view
[Next/Previous Months] [Main vserver Project Homepage] [Howto Subscribe/Unsubscribe] [Paul Sladen's vserver stuff]
Generated on Wed 04 Aug 2004 - 11:23:09 BST by hypermail 2.1.3