I debugged this problem and it was a spurious bug in how --paranoia was implemented, and librsync/rdiff both work fine without the --paranoia feature enabled.
This feature existed to catch and help debug potential rollsum bugs. The rollsum code is now very mature and the --paranoia feature itself has become a source of more bugs than the rollsum it was designed to check. So this problem has been fixed upstream with the release of v2.1.0 by removing the --paranoia feature. On Thu, 15 Aug 2019 at 21:35, Donovan Baarda <a...@minkirri.apana.org.au> wrote: > Thanks for reporting this. > > This should not be happening. It's unlikely to cause any corruption (which > would require a strong-sum collision), but it might be indicating a rollsum > bug that could be causing delta calculation to miss some matches, resulting > in a larger delta than it should. > > I was just considering removing the --paranoia setting because I thought > the delta code in librsync was pretty mature and we didn't need it any more. > > I filed an upstream bug against librsync and will investigate/fix; > > https://github.com/librsync/librsync/issues/155 > > -- > Donovan Baarda <a...@minkirri.apana.org.au> > -- Donovan Baarda <a...@minkirri.apana.org.au>
