Peer to Peer instead of Client to Server
john at betelgeuse.us
Mon Apr 7 16:07:35 EDT 2008
Douglas E. Engert wrote:
> Jeffrey Hutzelman wrote:
>> --On Monday, April 07, 2008 01:30:42 PM -0500 "Douglas E. Engert"
>> <deengert at anl.gov> wrote:
>>> This sounds like user2user. DCE had it and Windows has had it for
>>> years. There where some IETF Kerberos and GSSAPI drafts written
>>> by Microsoft, but never caried forward. Globus could do it through
>>> GSSAPI, using its GSI and there where mods to Kerberos to support
>>> user2user so Globus could call GSSAPI/Kerberos.
>> Kerberos does U2U; see RFC4120 section 3.7 for details.
> But is it in any GSSAPI implementations yet?
>> However, it's not clear to me that this application requires U2U, the
>> main feature of which is that it permits authentication to a "server"
>> which has a current TGT but does not know its long-term key. In the
>> described application, the servers all have keytabs and currently do not
>> run with tickets, so U2U really doesn't apply here.
> You are right, reading closer, it looks like U2U is not needed, as each
> server has a keytab.
> The key to the misunderstanding in the original note might be in:
> >> The server to server authentication is a bit trickier, though
> >> something like requesting a TGT for the server principal, then
> >> using that to get server tickets in the same way as a client
> >> does, except that the very begining of that process requires
> >> the input of a password, and If I go storing passwords on the
> >> servers, I might as well use the service password, yes? In other
> >> words, is there any way to use the server keytable as a means for
> >> getting a TGT?
> kinit -k -t can use a keytab to requests a ticket. No password is needed.
So it seems like the consensus is that some kind of server to server
authentication would be the way to go, using the server's own keytab
to request a ticket for any one of the other servers in the distributed
service (which has to be distributed, as the distributed service is
a multi-host storage area network, and the service being requested
is: "Tell me who out there can satisfy this storage constraint set?").
That sounds right to me from the replies I've gotten . . . thanks
for the input and the time!
More information about the krbdev