Re: [VIPR] vipr-overview-06

[Michael Procter <[e-mail]>]

Hi Jon,

Sorry for leaving comments until the last moment.  I've mainly read
section 7.5, as that seems to be the main area of change.

The main attack that doesn't seem to be mentioned is the one that
allows the approximate call start/end times to be used in conjunction
with multiple registrations in the RELOAD DHT to find accurate call
start/end times.  A malicious server can then use these to
authenticate itself as the caller to the called party, and perform a
MITM attack on all signalling and media for future calls.  This is
described in section 2.2 of draft-procter-vipr-privacy-concerns-00, in
the context of pranks, but it is certainly more widely applicable with
an increased number of registrations.

The reason I feel this attack should be included is that using call
start/end times is likely to play a part in any similar effort in the
future, and this is one of the attacks we should learn from.  A number
of the other attacks listed are likely to be avoided simply by
increased awareness of the risks of 'pervasive passive monitoring',
but this one is more active and should be noted for future attempts.

Best regards,

Michael