Re: [Tails-dev] Why doesn't Tails use tlsdate? (htp replacement)

2012-06-06 Thread intrigeri
Hi,

pro...@secure-mail.biz wrote (06 Jun 2012 01:15:55 GMT) :
> Why doesn't Tails use tlsdate, made by Jacob Appelbaum? [1] [2]

> Wouldn't it be a good replacement for htp?

Because it's not clear at all where, and how it could fit into,
replace or improve the current time sync' system we already have:
https://tails.boum.org/contribute/design/Time_syncing/

I had difficulties communicating on this topic with Jacob on IRC, so
I told him I will have a serious look once he makes this point clear.

tlsdate could be a good replacement for HTP once it has the features
we need (e.g. our three-pools design) -- OTOH, another options could
be to keep our existing htpdate wrapper (that implements the clever
bits), and merely replace wget + header parsing in there with tlsdate.
In this context, tlsdate communication would go through Tor.

However, it *seems* to me Jacob was suggesting us to run tlsdate in
the clear, that is without going through Tor; hence the question
I raised about the network fingerprint of this tool, unanswered as of
today (see <85sjgwz3kw@boum.org> on tor-talk). I'm worried running
tlsdate in the clear would get us back to the "Tails leaves a clear
bootstrap network fingerprint" old days we have managed to escape with
our current time sync' system.
___
tails-dev mailing list
tails-dev@boum.org
https://mailman.boum.org/listinfo/tails-dev


Re: [Tails-dev] Why doesn't Tails use tlsdate? (htp replacement)

2012-06-05 Thread Maxim Kammerer
My take on tlsdate:

On Wed, Jun 6, 2012 at 4:15 AM,   wrote:
> Wouldn't it be a good replacement for htp?

No, since tlsdate has no features: see TODO items 5 and 6, for
instance (daemonization and clock skewing), and also items 9 and 1
(proxy support and leaking local clock). The claim about “parsing the
header with questionable code”, on the other hand, is silly — see my
reply to the email that you referenced. Granted, I wrote about C code,
and Tails uses Perl HTPDate version, so YMMV. Even without https
support in C HTPDate, it seems more attractive to me than current
tlsdate. With all that said, I actually intend to fork tlsdate at some
point and implement the required features, but it's quite low
priority.

-- 
Maxim Kammerer
Liberté Linux: http://dee.su/liberte
___
tails-dev mailing list
tails-dev@boum.org
https://mailman.boum.org/listinfo/tails-dev


[Tails-dev] Why doesn't Tails use tlsdate? (htp replacement)

2012-06-05 Thread proper
Why doesn't Tails use tlsdate, made by Jacob Appelbaum? [1] [2]

Wouldn't it be a good replacement for htp?

[1] https://lists.torproject.org/pipermail/tor-talk/2012-February/023275.html
[2] https://github.com/ioerror/tlsdate

__
powered by Secure-Mail.biz - anonymous and secure e-mail accounts.

___
tails-dev mailing list
tails-dev@boum.org
https://mailman.boum.org/listinfo/tails-dev