Re: Weird-looking circuits in Vidalia

2008-03-26 Thread Robert Hogan
On Tuesday 25 March 2008 21:05:49 you wrote:
snip


 Ok, thanks for the info! How about replacing these strings with text like
 Directory Request in future? That would be be little more descriptive.

I was thinking the same thing recently. I even went so far as to start a 
proposal  - because there are numerous tunneled requests in Tor that aren't 
user 
initiated these days. I didn't get very far with it, and I'm not sure it's 
particularly straightforward in all cases. But it looks to be easy enough for 
these tunneled requests.

Here's what I was thinking:

Motivation/Overview:

  Tor now tunnels a large number of network maintenance operations through
circuits on the Tor network. Many of these operations are not initiated by
the user. Both TorK and Vidalia display active connections to the user and
these maintenance operations may cause alarm, distress, and even panic if
displayed without at least some attempt at explanation. If Tor were to provide
a STREAM_PURPOSE string as an extension for the existing STREAM_EVENT
controllers would be able to determine whether to display a stream to the
user, or more likely provide a mechanism for explaining the purpose
of the connection to the curious user.

Specify a new PURPOSE field for extended stream events as follows:

Index: doc/spec/control-spec.txt
===
--- doc/spec/control-spec.txt   (revision 14111)
+++ doc/spec/control-spec.txt   (working copy)
@@ -984,6 +984,7 @@
   650 SP STREAM SP StreamID SP StreamStatus SP CircID SP Target
   [SP REASON= Reason [ SP REMOTE_REASON= Reason ]]
   [SP SOURCE= Source] [ SP SOURCE_ADDR= Address : Port ]
+  [SP PURPOSE= Reason]
   CRLF

   StreamStatus =
@@ -1033,6 +1034,13 @@
that requested the connection, and can be (e.g.) used to look up the
requesting program.

+  Purpose = DIR_FETCH / UPLOAD_DESC / DNS_REQUEST /
+ USER /  DIRPORT_TEST
+
+   The PURPOSE field is provided only for NEW and NEWRESOLVE
+   events, and only if extended events are enabled (see 3.19).  Clients MUST
+   accept purposes not listed above.
+


signature.asc
Description: This is a digitally signed message part.


Weird-looking circuits in Vidalia

2008-03-25 Thread ZiM
Hello

When I started up the newly installed Tor 0.2.0.22-rc r14114 for first time, in 
addition to the typical Guardnode1,Nodename2,Nodename3 circuits, some weird 
lines ones below appeared in the Vidalia circuits list.

$F5EE4F21D...
$38014A01C...

I copied only first few characters to this message. Those look like raw Tor 
node fingerprints or something. They disappeared after Tor had been running for 
a while. Is this some new feature, Vidalia bug or something weird going on?


Re: Weird-looking circuits in Vidalia

2008-03-25 Thread Roger Dingledine
On Tue, Mar 25, 2008 at 01:41:51PM -0700, ZiM wrote:
 When I started up the newly installed Tor 0.2.0.22-rc r14114 for first
time, in addition to the typical Guardnode1,Nodename2,Nodename3 circuits,
some weird lines ones below appeared in the Vidalia circuits list.
 
 $F5EE4F21D...
 $38014A01C...
 
 I copied only first few characters to this message. Those look like raw
Tor node fingerprints or something. They disappeared after Tor had been
running for a while. Is this some new feature, Vidalia bug or something
weird going on?

They're caused by a new Tor feature. Starting with Tor 0.2.0.22-rc,
Tor does encrypted directory requests rather than plaintext (http)
directory requests. These show up in the Vidalia circuit list as a bunch
of one-hop circuits, since in reality that's what they are.

I wonder if Vidalia should hide these one-hop circuits from the user? Or
maybe users in the future will get used to seeing them in the list?

--Roger



Re: Weird-looking circuits in Vidalia

2008-03-25 Thread ZiM


--- [EMAIL PROTECTED] wrote:

From: Roger Dingledine [EMAIL PROTECTED]
To: or-talk@freehaven.net
Subject: Re: Weird-looking circuits in Vidalia
Date: Tue, 25 Mar 2008 16:53:12 -0400

On Tue, Mar 25, 2008 at 01:41:51PM -0700, ZiM wrote:
 When I started up the newly installed Tor 0.2.0.22-rc r14114 for first
time, in addition to the typical Guardnode1,Nodename2,Nodename3 circuits,
some weird lines ones below appeared in the Vidalia circuits list.
 
 $F5EE4F21D...
 $38014A01C...
 
 I copied only first few characters to this message. Those look like raw
Tor node fingerprints or something. They disappeared after Tor had been
running for a while. Is this some new feature, Vidalia bug or something
weird going on?

They're caused by a new Tor feature. Starting with Tor 0.2.0.22-rc,
Tor does encrypted directory requests rather than plaintext (http)
directory requests. These show up in the Vidalia circuit list as a bunch
of one-hop circuits, since in reality that's what they are.

I wonder if Vidalia should hide these one-hop circuits from the user? Or
maybe users in the future will get used to seeing them in the list?

--Roger


Ok, thanks for the info! How about replacing these strings with text like 
Directory Request in future? That would be be little more descriptive.