On Mon, Nov 16, 2009 at 12:11:41AM +0100, Jelmer Vernooij wrote: > I'm not sure what the best thing is to do here. > > This situation should be unusual, and it does not corrupt any of the > data in the original Subversion or Bazaar repositories. It's only a > problem if you push introduce a new file in a merged revision that is > not changed by the merging revision and then try to pull those changes > down to a bzr repository that is older than 2a. Even after that, bzr > will complain but the data should be ok. > > Putting a newer version of bzr-svn into Lenny wouldn't really be an > option, as this would require us to put newer versions of bzr, bzr-gtk, > etc in as well. This would be too risky. > > I don't think removing bzr-svn completely would be necessary. Another > alternative would be to simply disable the ability to do round-tripping > pushes (or round-tripping pushes of merges?) into Subversion and only > allow dpush.
Hi Jelmer, Note that the second time we experienced repository corruption, we were using bzr-svn 1.0.0 and only for pulls, never for pushing to svn. This was with a pre-2a format though. I'm not sure what's the cause anymore. It could also have been related to our use of bzr+ssh (see https://savannah.gnu.org/support/index.php?107077) with an old (1.6) bzr daemon. -- Robert Millan The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and how) you may access your data; but nobody's threatening your freedom: we still allow you to remove your data and not access it at all." -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
