[Monotone-devel] Re: Hash collisions resiliency

[Frank Ch. Eigler]

Nathaniel Smith <address@hidden> writes:

> > [...]  Create a specially crippled version of
> > monotone, which uses (say) only 12-bit hashes.  Run whatever test
> > suite you can, a bunch of times.  See what blows up.  [...]
> 
> This is kinda a silly idea?  What will blow up is that at netsync
> time, monotone will simply not realize that it is missing some data,
> and silently replace the data it is supposed to fetch with data it
> already has.  There's no practical consistency check against this.
> [...]

Then perhaps the test suite could use a few additions, such as
explicit byte-wise comparisons of expected vs. actual stored files.
Instead of asserting that monotone return codes are zero, actually
archive and compare large bodies of text before/after version control
operations.  That is the primary observable from a user's point of
view anyway.


- FChE