[tor-dev] Entry guards, primary guards, dir guards

George Kadianakis desnacked at riseup.net
Mon Feb 8 21:48:23 UTC 2016

Ola Bini <obini at thoughtworks.com> writes:

> Hey,
> Thanks- this is very helpful.
> When it comes to vanguards, I've already read through the
> proposal. I'm not exactly sure I understand how much different 259
> would need to be to support the 247 needs. It seems we should be able
> to just run the algorithm NUM_SECOND_GUARDS * NUM_THIRD_GUARDS times
> to choose the sets of vanguards for each layer, right?

Hmm, how would that work exactly?

Let's say I'm a prop247 hidden service. I just received an introduction and
want to setup my rendezvous circuit.

To setup my circuit, I would need to do three guard picks, one for every
layer. Each layer has a different guard list.

First, I use my layer-1 guardlist to pick my layer-1 guard. That's easy, I use
a single guard for layer-1, so I always pick the first reachable non-bad guard
from the layer-1 guardlist.

Then I need to use my layer-2 guardlist to pick my layer-2 guard.  Proposal 247
says that each HS has two layer-2 guards , so I would need to pick a guard out
of the two top guards in my layer-2 guardlist. How does this happen exactly?

A similar thing needs to happen for layer-3.

More information about the tor-dev mailing list