Re: [rdiff-backup-users] Tar replacement - format proposal

[Ben Escoto]

On 27 Sep 2003 00:34:26 +0100 Kevin Spicer <address@hidden> wrote:
> Some thoughts on the subject of inodes, steming from posts by Will Dyson
> and John Goerzen, as well as Ben's webpage...
> 
> It seems to me there are two (semi-separate) issues related to inodes...
> 
> 1) The original inode numbers (from the source filesystem)
> 2) A system of inode numbers for the new file format so that it could be
> used accessed as a filesystem, given the appropriate kernel support.
    ...
> There are a number of reasons why the original inode number may be
> needed or useful...
    ...
> A couple of problems spring to mind.

Probably the best thing to do is to take Will Dyson's suggestion and
use the inner file offset of the file's metadata entry as its inode.
There would be no danger of collision, we wouldn't need an inode
table, and even in a large archive inode numbers would be less than 32
bits.

But that doesn't mean we can't find hardlinks and renamed files.  We
can record the original inode and device number also, under a separate
heading perhaps.  If we have a flexible metadata format there's no
need to be limited to the traditional metadata.


-- 
Ben Escoto

pgpkEZWtoQodm.pgp
Description: PGP signature