Date: Fri, 13 Feb 2004 15:15:53 +0100 From: Bernd Walter <ticso@cicely12.cicely.de> To: "Marc G. Fournier" <scrappy@hub.org> Cc: freebsd-hackers@freebsd.org Subject: Re: Subversion/CVS experiment summary Message-ID: <20040213141551.GF44313@cicely12.cicely.de> In-Reply-To: <20040212164236.D96890@ganymede.hub.org> References: <200402091130.05656.craig@tobuj.gank.org> <20040209185105.O72949@fw.reifenberger.com> <20040212164236.D96890@ganymede.hub.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, Feb 12, 2004 at 04:48:25PM -0400, Marc G. Fournier wrote: > note that this was pre-0.34, but since its still "under development", > there is always the chance that this happens again ... a load of the > system took 49hrs, I believe was mentioned ... how long to dump/reload the > system once its already "in subvsersion" format? dump/load is quite fast, but dump format includes the full content of each revision - piping to gzip befor storing is advised for big trees. dump format can easily used to distribute the tree by taking hourly snaps, but not for distribution of checkouts as CVSup can do. Unfortunatley general speed is still an issue. Try to commit a very large diff and you wait forever if you are not going out of memory bevor. There are also obscure workdir problems from time to time which could easily confuse the wide user base. -- B.Walter BWCT http://www.bwct.de ticso@bwct.de info@bwct.de
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040213141551.GF44313>