Tim Bader schrieb:
> I've been reading through the subversion book and the tortoise help
> files and I'm still quite confused about the different network
> protocols versus file protocols.
> I've been testing tortoise with a local repository in preparation for
> putting our main development server under source control.
> My local efforts work fine and tortoise seems to be using the file
> protocol automatically.
This depends on the kind of repository url set in the checkout dialog:
e.g. file:///V:/DOKUMENTATION/Repository/project1 -> starts with
file:/// => file protocol
svn://EENT540/project1 -> starts with svn:// => client-server
protocol (EENT540 is the server name in the network).
Using the client-server protocol needs a server software (apache or
svnserve running on the server). For setting up a server see
> Our dev server is on the LAN so can I simply install subversion on the
> server and continue to use tortoise "as is": it's not a mapped drive
> at the moment (but could be) but I can navigate there by the normal means.
> I even tried setting up a repository on the server and it worked fine
> even though subversion isn't installed there: is this because
> subversion is working on my local machine or just that tortoise
> doesn't need it?
I don't understand this. How can you access the drive on the server if
it is not mapped to your local machine ?
There are two possibilities in principle:
using the file protocol:
Map the server drive to your local machine.
Advantage: no difference to local behaviour
Disadvantage: no read access control to parts of the repository,
everybody can administrate the repository
Risk: repository can be deleted accidentally, repository is becoming
unuseable if you use a Berkeley database backend.
If you use the native file system backend this is no issue.
We use this configuration for about half a year and no problems occured.
using the client-server protocol:
You have to set up a server like svnserve or apache. Setting up svnserve
is very easy on a windows based server.
> Also, I've read through the stuff about authentication, etc and I
> think that for my situation we only need to know who the author of a
> change was and not to force them to enter a password, etc.
When using the file protocol, no password is needed. TSVN uses the
windows user name as author. When using the client-server protocol you
can give anonymous access without a password, too. Doing this way
enables everybody in your network to access the repository - without
firewall maybe the whole world.
> -Somehow, tortoise seems to have picked up my name automatically as
> the author for all commits I have been making, but I haven't
> configured anything to tell it that: how does it "know" + will it work
> ok when I've got different users accessing the repository from their
> machines (They will all have local workspaces except for one who for
> technical reasons will need his own workspace on the server)?
> I suppose what I'm working up to is whether I need to set up a server
> (either Apache or svnserve) or if I can just install tortoise on each
> local machine.
> Sorry, that's about 3 questions in one!
> Thanks in advance.
> Tim Bader
> Mailto: firstname.lastname@example.org <mailto:email@example.com>
> Telephone: 01932834 829
> This message has been checked for all known viruses.
> Its content and any files transmitted with it are intended solely for
> the addressee(s) and may be legally privileged and/or confidential. If
> you are not the intended recipient please delete and contact the
> sender shown above via email return.
> Unless expressly stated otherwise this email does not form part of a
> legally binding contract or agreement between the recipient and
> Euroking Miracle Ltd.
# Dipl.-Ing. Sven Probst
# Fraunhofer Institut f. Lasertechnik
# Steinbachstr. 15, D-52074 Aachen, Germany.
# Tel : + 49 (0) 241 8906 314
# Fax : + 49 (0) 241 8906 121
# EMail : firstname.lastname@example.org
# WWW : <http://www.ilt.fraunhofer.de>
Received on Wed Aug 2 14:23:40 2006