On 23 Aug 2006, at 20:53, Nico Kadel-Garcia wrote:
>
>
>>
>> On 23 Aug 2006, at 11:30, Nico Kadel-Garcia wrote:
>
>>> Well, true. It's most easily enforceable on machines where I'm
>>> the sys-admin, and could put my copy in /usr/bin or /usr/local/bin.
>>
>> PATH=$HOME/unauthorisedsvn/bin:$PATH
>> export PATH
>> svn ....
>
> Again, true. They could also create a .login message that says "My
> Password is: XXXX".
>
> But there's a practical distinction between leaving the passwords
> lying around by default in a well-known, accessible space (and
> having permissions of 700 is still pretty accessible, especially in
> an NFS home directory environment or a laptop or unencrypted backup
> environment!), and in leaving user's enough room to hang themselves
> if they really insist.
>
Talking specifically about your modified svn which does not store
passwords, an effect of that would be your users having to type their
passwords in lots of times which some of them might find irritating.
It's quite conceivable that some of them would discover the
subversion web site and download the unmodified software. It's far
less likely that they would deliberately echo their password to the
terminal whenever they logged in.
You could, of course, change the client config file to not store
their password and then explain to your users why you have done it.
That'd probably stop them from changing it back.
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Thu Aug 24 12:00:46 2006