On Sat, Feb 05, 2005 at 06:19:07PM -0500, William Allen Simpson wrote:
>
> How about one of the newer attempts, subversion?  I recently had a
> project with it, but since I was only using client side, I have no
> idea what the server side is like.

We experimented with it: one of the NetBSD developers spent a lot of time
fixing cvs2svn so it would import our repository _at all_ without crashing,
and then discovered that the resulting Subversion repository was four times
as large as the original CVS repository had been.

In other words, we'd need a machine with 32GB of memory to keep it in RAM.
Unfortunately, we don't have tens of thousands of dollars lying around to
devote to such experiments. :-/

We do have some changes to cvs from Christian Limpach that make it work
better in our environment: we run a modified cvs on the master anoncvs
server that keeps temporary files/directories in a DB database rather
than in the filesystem, and another set of changes we don't use yet that
add support for changesets.  The main problem is just that the repository
grows as we pull more software like X under our own version control, and
as pkgsrc grows, so we end up needing more RAM to cache it all.  But
that's in the works.

-- 
 Thor Lancelot Simon	                                      tls@rek.tjls.com

"The inconsistency is startling, though admittedly, if consistency is to be
 abandoned or transcended, there is no problem."		- Noam Chomsky