On 12/29/05, Dr. Stephen Henson <[EMAIL PROTECTED]> wrote:
> On Thu, Dec 29, 2005, WebSpider wrote:
>
> > The issue I'm running into is as follows: A new node (A) is about to
> > make it's first connection to an already existing node (B). The new
> > node knows the IP address and port number by use of a configuration
> > file.
> >
> > The already existing node (B) has the posession of the following data:
> >
> > * The public root certificate
> > * The full list of all signed certificates
> > * The CRL
> > * It's own public certificate and private key for the certificate
> >
> > The new node (A) has the posession of the following data:
> >
> > * The public root certificate
> > * It's own public certificate and private key for the certificate
> >
> > Some more information:
> > * Both the full list of signed certificates and CRL are not available
> > from any other source than the already existing node
> > * The commonName field is not to be used to identify the remote host,
> > since the value of the commonName field in the certificate of the
> > already existing node may vary
> >
> > How can I make the new node (A) send an encrypted request to the
> > already existing node (B) while node A does not have any public
> > key/certificate information about the already existing node (B), and
> > still make sure that I am actually talking to B, and not some
> > Man-In-The-Middle ?
> >
>
> So from your description new nodes already have the public root certificate
> from some trusted means?

This is true.

> That at least means that a man in the middle attack can only be performed by
> an entity posessing a certificate issued by the root CA.

Good thing to know, thanks !

> That leaves the possibility of a new node (A') impersonating an existing node.
> To avoid that you need to be able to identify nodes securely.
>
> How you do that varies. For HTTPS websites for example the certificate
> contains details of the host name and the CA is implicitly trusted to check
> the validity of the entity claiming to represent that host name before issuing
> a certificate.

Right. Then I guess that is something I need to look at, since I
currently cannot reliably identify a remote node, and therefore seem
to be lacking the public key information to initiate secure
communication.

Would you happen to know other ways of identifying a remote node,
other than to trust DNS?

Greetings,

Nils
--
Simple guidelines to happiness:
Work like you don't need the money,
Love like your heart has never been broken and
Dance like no one can see you.
______________________________________________________________________
OpenSSL Project                                 http://www.openssl.org
User Support Mailing List                    [email protected]
Automated List Manager                           [EMAIL PROTECTED]

Reply via email to