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

Re: svn commit: r19451 - trunk/subversion/tests/cmdline

From: Alan Barrett <apb_at_cequrux.com>
Date: 2006-04-28 08:54:34 CEST

On Wed, 26 Apr 2006, John Peacock wrote:
> Alan Barrett wrote:
> > I don't understand this comment. In what sense does a user of
> > tunneled svn need write access to the repository files?
>
> In the sense that svn+ssh:// is exactly the same as file://, in terms
> of requiring write access to the repository files.[1]

All of svn+*://, svn:// and http[s]:// requires write access for the
user used by the web server or svnserve on the server side, but do not
require write access for the user on the client side.

file:// has the same user on both the client (working copy) and server
(repository) side, and requires write access for that user. This is
very different from the others.

> 1. Yes, you can run svn+ssh:// using a single shared user (via the
> command= parameter of the authorized_keys file under .ssh), but this
> still requires the user actually accessing the repository to have
> write access.

Yes, the user on the server side needs write access. That's the same
for any access method. The interesting thing is how much access the
user on the client side has, and svn+ssh allows the client user to be
given no access at all except through the mediation of svnserve.
If svnserve uses authz for plain svn:// access, it would be very
surprising for it not to do so for svn+ssh:// access.

--apb (Alan Barrett)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Fri Apr 28 08:55:34 2006

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

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