[TICTOC] security ID submitted for review

["David L. Mills" <[e-mail]>]

Guys,

Here are some general comments on the ID.  I apologize if they are a 
little rough as I have to dictate this message instead of my usual 
polished wordsmithing.

format.
The ID does not specify an extension field format.  Presumably, the 
extension field format described in RFC 5906 is adequate with new 
message types as required.  However, in keeping with long Internet 
tradition, it must be possible to determine correct format without 
interpreting the header or extension fields.  In addition, it must be 
possible to verify message integrity without interpreting the extension 
fields.  In particular, the private key for the message authentication 
code (MAC) , if used, must be known beforehand.  This is where the 
"cookie snatcher" issue comes up as described in previous messages.  In 
particular, an important constraint for normal operation is that 
extension fields are not used unless required for cookie determination.

Glaring vulnerability exists when using extension fields, as the key 
value is public and must be known in advance.  This invites a 
cut-and-paste attack, as well as header changes, even if subsequently 
detected by the on-wire protocol.


certificate.
Both RFC 5906 and the ID require the server certificate.  The server 
certificate can be conveyed, either off-line, or directly or indirectly 
via a certificate trail as described in RFC 5906 or the ID.  If obtained 
via the trail, there is a hazzard due to a middle-man attack and 
masquerade.  A solution proposed in RFC 5906 uses an identity scheme 
such as IFF.  There may be an equivalent, or better schemes available now.

protocol.
Most of the discussion following the ID announcement is from protocol 
weenies, affectionately known as protomeister.  When the Autokey scheme 
was first defined in 1996,  the signature time required ranged up to one 
second on the computers of the time.  Therefore, a clogging hazzard 
existed in the form of  a replay attack.  This is why the timestamp was 
included in the extension field format to suppress duplicates.  Most of 
the extension fields were computed and signed off-line.  Nevertheless, a 
clogging vulnerability remains in the cookie request message.  This 
requires the server to do two public key incriptions and two hash 
functions. Today, this overhead may no longer  be considered  excessive; 
however, this may not be true for public time servers with thousands and 
thousands of clients.

In broadcast mode, the protocols described in RFC 5906 and the ID are 
messy and with several possible vulnerabilities.  Here is a suggestion 
that might be useful.  Define a new broadcast packet format as the NTP 
header followed by one extension field representing the MAC.  The 
extension field represents the signature of the header computed using 
the private key of the server and can be verified by the client using 
the public key in the server certificate.  Originally, this was 
considered an unacceptable dilution of accuracy due to the variability 
of the signature computing time and the length of the signature.  
However, at least in modern NTP, the interleaved broadcast mode can be 
used to correct for this overhead.  This assumes that the overhead for 
either the server or client is reduced using fast modern processors.

As mentioned previously, in NTP symmetric mode the current protocol in 
RFC 5906 should be changed to use Diffie-Hellman key agreement.  This is 
so easy to do that it can be incorporated separately in the standard.

While the interleaved mode is useful in symmetric and broadcast modes, 
it can't be used when the server does not maintain state separately for 
each client.  However, as in the orginal NTP design many years ago, the 
output delay can be computed for each transmitted server packet.  The 
output delay can be averaged and used to augment  the transmit timestamp 
or retrieve using an extension field.

Note that this message does not address the cookie snatcher issue 
discussed by our German protomeisters, who have concocted a brillient 
solution.  Many of the issues discussed are relevant to both NTP and 
PTP, although the on-wire protocol details may be different.  I hope 
this discussion is useful in either context.

Dave