Danek Duvall wrote:
> In particular the first hash may not match what's on disk, because we
> decide to update a file based on its elfhash, if the attribute exists.
> Many ELF files will be rebuilt even if the source didn't change, and things
> like the comment section will change, but there's little point in
> retrieving that file, since there will be no difference in execution.

While I agree that at a technical level this is probably fine, it seems 
like it's setting us up for support problems.  Users and technicians 
alike will look at things like file sizes, file checksums, and file mod 
times, and will ask questions about why their two installations don't 
match.  There's an answer that might well satisfy them, but the mere 
fact that they asked the question costs us time and customer confidence.

If we believe that there are parts of ELF files that (a) routinely 
change and (b) don't matter, and we want to avoid delivering unnecessary 
change to the customer, perhaps we should do that filtering at an 
earlier point, so that we never even deliver the change internally... 
perhaps all the way back to not including those sections in the file in 
the first place.
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss

Reply via email to