Re: [tor-talk] Tor 0.2.3.2-alpha is out

[Roger Dingledine <arma@xxxxxxx>]

On Wed, Jul 20, 2011 at 11:15:54PM -0600, Jim wrote:
> So far, after a brief test, everything seems to be fine.  I was curious
> about what was logged and wondering if it represents any kind of an
> error or bug.  The full log is listed at the end of this email.  The
> lines I am wondering about are:
>
> Jul 20 19:42:30 Host Tor[9506]: I learned some more directory
> information, but not enough to build a circuit: We have only 0/2480
> usable microdescriptors.
> Jul 20 19:43:22 Host Tor[9506]: We now have enough directory information
> to build circuits.
> Jul 20 19:44:10 Host Tor[9506]: Bootstrapped 100%: Done.
>
> I am wondering if the first line is accurately reporting the state when
> 52 seconds later (with a 52kbps connection) it reports it has enough
> information to build circuits and 100 seconds after which, it reports
> 100% bootstrapped and stops downloading data (observer using iftop).

What you describe is plausible. If you want to get more confidence, you
could rm -rf ~/.tor/cached-* so your Tor client has to fetch everything
from scratch.

The "we now have enough" line shows up once Tor has at least 1/4 of
the descriptors (or in this case microdescriptors) that it knows it
wants. So it makes sense that it takes a while after that point to a)
fetch the rest of them, and b) build a circuit.

Note that the main benefit you'll see from microdescriptors is in reduced
directory overhead *after* initial bootstrap -- because microdescriptors
don't change as often as normal descriptors so you can keep them cached
for a week. They don't help so much with the initial bootstrapping cost.

--Roger

_______________________________________________
tor-talk mailing list
tor-talk@xxxxxxxxxxxxxxxxxxxx
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk