>>>>> "Simon" == Simon Burge <simonb@telstra.com.au> writes:

Simon> Maybe storing a checksum in the di_spare field of the inode is
Simon> a solution to this.  Two problems with this are:

[...]

Simon> An ulgy idea is to store the file's checksum in the _next_
Simon> file's inode data, and have a dummy inode at the end of the
Simon> tape with the last file's checksum in it.

Simon> Sound good, bad or otherwise?

It's esthetically very ugly, but seems reasonable from a practical
standpoint.  In fact, that's how the amanda folks want to
(eventually) implement their checksums, only s/inode/file on tape/g

Chris

-- 
-----------------------------------------------------cjones@math.montana.edu
Chris Jones                                          cjones@honors.montana.edu
           Mad scientist at large                    cjones@nervana.montana.edu
"Is this going to be a stand-up programming session, sir, or another bug hunt?"