[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: a solution on the horizon to counteract detecting temperature through clock skew (on Tor Hidden Services)?

On 5/14/07, Wes Felter <wesley@xxxxxxxxxx> wrote:
... Power analysis is not the
same thing as temperature-induced clock skew.

i've wondered about using a frequent ntpdate to reduce skew, and if that is not sufficient, what about a modified client that uses adjtime() or settimeofday() with random offsets (+/- within some tolerance) to gently randomize on a few minute schedule?

is there a public tool to analyze clock skew like that mentioned in
the attacks to determine if such workarounds are effective?

best regards,