[EMAIL PROTECTED] on 06/08/2000 12:13:35 PM
>Noel L Yap writes:
>>
>> Client/server CVS can easily map users if the client sent the remote user's
name
>> over to the server to use in its logs. Therefore, I propose that such an
>> enhancement should be done (I'll even do the work if my schedule doesn't get
>> overloaded).
>
>The main problem I see with this is that you lose all accountability --
>the client can claim to be anyone and the server will just blindly
>accept that the client is telling the truth. pserver does at least some
>authentication by requiring that the user be known and requiring the
>corresponding password.
Pserver authentication is extremely bogus. Better authentication can be done
using SSH (if people really cared about it). If people didn't care too much
about it, they'd rely on client login as enough authentication.
If people are really against such a change, another $CVSROOT/CVSROOT/config
parameter (eg UseRemoteUserName) can be created. I wouldn't even mind too much
if the default were "no", but I think setting to "yes" can allow better audit
trails (ie better security) in some situations. For example, given that someone
had gotten hold of someone else's private SSH key and used it to commit
something, their (ie the hacker's (the term "hacker" used extremely broadly
here)) username will get recorded. (I also think that a more general solution
to the audit trail problem would be if SSH set an environment variable (eg
REMOTE_USER) but that's another proposal ;)
Noel
This communication is for informational purposes only. It is not intended as
an offer or solicitation for the purchase or sale of any financial instrument
or as an official confirmation of any transaction. All market prices, data
and other information are not warranted as to completeness or accuracy and
are subject to change without notice. Any comments or statements made herein
do not necessarily reflect those of J.P. Morgan & Co. Incorporated, its
subsidiaries and affiliates.
