[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Several Questions

From: Ryan Schmidt <subversion-2006Q1_at_ryandesign.com>
Date: 2006-02-10 20:10:53 CET

On Feb 10, 2006, at 09:01, Christoph Trappl wrote:

>>> - Can I use SQL, MSDE or Oracle instead of the Berkley DB? (if yes,
>>> how?)
>>
>> No, the only available backends are BerkeleyDB and FSFS, a custom
>> format designed just for Subversion. FSFS is available since
>> Subversion 1.1.0 and is the default type for new repositories since
>> 1.2.0 and is recommended. For a rundown of the differences between
>> BDB and FSFS see here:
>>
>> http://svn.collab.net/repos/svn/trunk/notes/fsfs
>
> SO FAR, SO GOOD (OR NOT:-)).

What do you mean, "or not"? Why would you care how Subversion keeps
your data safe? Isn't it enough to know that it does?

I believe it was just last month that there was a discussion on this
mailing list about what benefits people percieved by using an SQL
backend. You can look it up in the archive if you like, but suffice
it to say the topic has been brought up several times before, but in
the end, the answer is that Subversion can't do that right now, and
it's not a developer priority right now because it's not clear what
the benefits are over the current implementation, or that the
benefits, if there are any, justify the development effort, when
perhaps the benefits can be achived in other ways, sometimes through
other tools which already exist.

> CITRIX IS IN GENERAL USED IN COMBINATION WITH TERMINAL SERVERS,
> MEANING ALL
> CLIENTS START THEIR PROGRAMS CENTRALLY ON A SERVER. THUS SOME MORE
> USERS
> MIGHT HAVE SIMULTANEOUS ACCESS TO CERTAIN SERVICES / TASKS.

> NORMALLY, WHEN WORKING WITH A TERMINAL SERVER, ALL PRORGRAMS, DATA,
> ETC. IS
> KEPT CENTRALLY ON THE SERVER. THIS ALSO APPLIES, THAT LOOKING AT
> SUBVERSION
> LOCAL WORKING COPIES ARE MIGHT BE PERMITTED, BUT I AM NOT SURE.
> JUST WANTED
> TO KNOW IF ANY EXPERIENCE, OR EVEN BETTER A HOW TO KIT EXISTS.

I still can't quite get a mental picture of what that really means. I
can tell you that most people have a repository server and then have
working copies on their local machines, do work, test it, then check
it in when it works. This works well for many kinds of work, such as
compiled software applications or non-programming collaboration. In
our company, we develop PHP/MySQL web sites, so we happen to put our
working copies on a central server which is set up with Apache and
MySQL, and our users access those working copies either via a Samba
mount on their Windows machines, or via an SSH session to the server
over the command line.

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Fri Feb 10 20:15:22 2006

This is an archived mail posted to the Subversion Users mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.