February 2014 - tor-relays - lists.torproject.org

Re: [tor-relays] TTTT, multiple runs
by Sebastian Urbach 19 Feb '14

19 Feb '14

Hi Sanjeev, Basically yes. Here are the details: http://web.engr.illinois.edu/~das17/tor-traceroute_v1.html#q-canrunelsewhere Feel free to ask if anything is unclear. -- Mit freundlichen Grüssen / Sincerely yours Sebastian Urbach ---------------------------------------------- Those who would give up essential Liberty, to purchase a little temporary Safety, deserve neither Liberty nor Safety. ---------------------------------------------- Benjamin Franklin (1706 - 1790), Inventor, journalist, printer, diplomat and statesman On Wed, Feb 19, 2014 at 6:53 PM, Sebastian Urbach <sebastian(a)urbach.org>wrote: > Right now we are at 95 ip's with at least 1 completed run. Hopefully we > can add another 5 to that number and reach the "100 milestone" before the > next data review in 03/2014. Multiple run's from the same ip are very > welcome as well. > Sebastian, Does this have to be run on a Tor Relay? I thought so, for my first run, but re-reading it now I am not sure. If running this on a non-relay would help, I can add a few more servers tonight. -- Sanjeev Gupta +65 98551208 http://www.linkedin.com/in/ghane

1 0

Enable statistics for hardcoded obfsbridges
by George Kadianakis 19 Feb '14

19 Feb '14

Hello people, I'm sending you an email because your bridges are hardcoded in the pluggable transport bundles. We recently added PT-specific metrics in metrics.torproject.org Specificaally, you can see a graph of PT users here: https://metrics.torproject.org/users.html?graph=userstats-bridge-transport If you select the obfs2 transport, you can see that it displays only 300 users, which is not entirely true. We suspect that the reason for the low number of users is because not many bridges are reporting PT statistics to the bridge authorities. This might be for two reasons: a) Your bridge might be configured to be private, that is to not publish its descriptor. This might happen if you have 'PublishServerDescriptor' set to 0 in your torrc. If you have a bridge hardcoded in the bundles, not publishing your descriptor doesn't really provide any security (your bridge is exposed anyway), so you might as well start publishing your descriptor, since this will allow us to gather statistics about the number of users you see. b) Your Tor might not be bleeding edge: Tor only started reporting PT statistics to the auths in tor-0.2.5.1. To upgrade to this version of Tor you can use either the git master or the automated Debian package builds (see https://lists.torproject.org/pipermail/tor-relays/2014-February/003899.html for how to use them). c) You don't have Extended ORPort enabled. With recent versions of Tor, you can enable it by adding 'ExtORPort auto' to your torrc. It would be great if you could fix the above so that your Tor can start submitting statistics! If you have any questions, feel free to ask me!

1 0

Working init script for TOR 2.5.1
by tor＠afo-tm.org 17 Feb '14

17 Feb '14

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Hello, i search a working init script for multiple tor instances. I tried it from https://www.torservers.net/misc/config/initd-tor but that is broken. Does somebody have a working one? Thanks for help -----BEGIN PGP SIGNATURE----- iQIcBAEBCgAGBQJTASW0AAoJEKawimmefC2InIkP/0guExgToj+B/hk9cQouPNdn J9ml+qdZxeEJH2G8V5NFhre4L1y4lubr9DhvVN0eMQ/vU+Dd/f0KqpO/05DD2pMG z2MJxq5qfvL4M/BApGbsC/+xIZBVRWHNHywO/hlR89sXn1Yd30kbnXYlAIQNCobA 6xsT4zhZ+crhKzUgC6jouIdzq1Wf9nfiChxf4w9UgCbO+t6JTkRBjyX7w0bHcTAP e3hMZBUEiJAuWeB3vubQPiQwvhrqw87yvSj6uNhKAq5u8nWtRo1mStD69jia+bdB yZIKxRgc8Y/+T/Ls8zWDiRvBUdvFd0buVQKiWjn8dX4UNoZNuFGIGtv5U+PfKP+3 osBmf3v20VmkW/7RcC/TAO0/9kqDupkdF4/W9OhzcIKQMv64OwJiEIFjE6mtTM8D sHhRbk++3YsVhSCy504KUvohewltEMpHpGD+/1vDwfebryeelBZTkn6xr51Kk6IM LijzNdcZFwmEtP5NbuioF07NyN5qKMyGUYdUhiDc51BT5ynkis4950lHMgJkpKrO 5ssv1Kw+f3B+53s67LbbaaPwErk1AdxxFnIBZ8RcDMLJIk3BsT/WwujPlF/Ftog1 cc6jLCHcxalMPzBT9b4NbNz9EbpOhqpQE/oUeasBtfPr9606bN2EHSTey4WVcuPp kLPxVVMrvY1B4W4fxI/Q =dy9S -----END PGP SIGNATURE-----

3 4

Tor Relay Downloading much more data than uploading
by borgersnick＠gmail.com 17 Feb '14

17 Feb '14

The log entry is: Heartbeat: Tor's uptime is 23 days 0:00 hours, with 0 circuits open. I've sent 6.83 MB and received 177.12 MB. I don’t think I made any changes to the torrc, intend to provide a relay. I am using trickle to limit bandwidth consumption instead of the built-in method because I didn’t want to learn about the bursting concept. ExecStart = /usr/bin/trickle -d 500 -u 500 -t .5 /usr/bin/tor --runasdaemon 0 --defaults-torrc /usr/share/tor/defaults-torrc -f /etc/tor/torrc --quiet Running on Fedora. Firewall allows external port 80 and 443 access to the server.

1 0

Dynamic upload limit
by Bram L 17 Feb '14

17 Feb '14

Dear list members, I am running a tor relay, but I have an upload limit (50GB/7 days). This limit usually can't be increased too much, because that might affect other network users. I would like to request my provider (university) for a bandwidth-free connection during the night (or whenever the network is not used very much), and a limited version during the peak times. My idea is that this way I could contribute to the Tor network without disturbing other users of my university network. Before I do so, however, I am curious whether it actually contributes to the Tor network. I noticed it takes time for the bandwidth to increase, and if my daytime upload limit is low (e.g. 80 kb/s) and unlimited (Gb/s) at night, my overall contribution could be very limited due to the time it takes Tor to utilize my available bandwidth. Do you have any insights in this matter? Thanks in advance! PS: currently, I drain the 50GB in about a day unlimited or I run at very slow speeds... Both don't give me the feeling I really contribute.

2 1

Re: [tor-relays] TTTT project
by Sebastian Urbach 15 Feb '14

15 Feb '14

Hi Mateusz, 19 days with 1/5 of the default pps value (1000) That would be around 4 days with pps 1000. That's exactly the value estimated in the project faq. Everything seems to be fine. Project site / faq: http://web.engr.illinois.edu/~das17/tor-traceroute_v1.html#faq Results: http://datarepo.cs.illinois.edu/relay_scoreboard.html Thank you for participating ! -- Mit freundlichen Grüssen / Sincerely yours Sebastian Urbach ---------------------------------------------- Those who would give up essential Liberty, to purchase a little temporary Safety, deserve neither Liberty nor Safety. ---------------------------------------------- Benjamin Franklin (1706 - 1790), Inventor, journalist, printer, diplomat and statesman On 13 Feb 2014, at 13:10, Sebastian Urbach wrote: > Dear honorable list members, > Some stats from my side: bash-3.2$ PPS=200 DONTERASE=yes time ./traceroutes.sh 1645980.86 real 28760.92 user 181530.51 sys Started on 27 Jan 2014 00:49 GMT Finished on 15 Feb 2014 02:02 GMT ca. 19 days to complete with scamper on OSX 10.7 Thanks, -mateusz

1 0

TTTT project
by Sebastian Urbach 15 Feb '14

15 Feb '14

Dear honorable list members, If anybody has questions regarding his / her submitted results than please include your public ip address in your mail. We can process your requests way faster if we have this essential information and don't have to ask for it. Btw. we are at 88 ip's with at least 1 complete run. Come on, let's make it 100 before the data review in March :-) Thank you very much for your attention ! The project: http://web.engr.illinois.edu/~das17/tor-traceroute_v1.html The results: http://datarepo.cs.illinois.edu/relay_scoreboard.html -- Mit freundlichen Grüssen / Sincerely yours Sebastian Urbach ---------------------------------------------- Those who would give up essential Liberty, to purchase a little temporary Safety, deserve neither Liberty nor Safety. ---------------------------------------------- Benjamin Franklin (1706 - 1790), Inventor, journalist, printer, diplomat and statesman

2 1

obfs3 a risk to scramblesuit?
by Delton Barnes 14 Feb '14

14 Feb '14

I am running a bridge with both obfs3 and scramblesuit. obfs3 is vulnerable to active probing. Say someone in China makes the mistake of connecting with obfs3 instead of scramblesuit, and active probing identifies it as a Tor bridge. Wouldn't the bridge IP address then be blacklisted, making it impossible for other people to connect even with scramblesuit? Would it make sense, then, as a bridge operator to elect to only support scramblesuit, even though no one will use it for a while? It seems you are only as good as your weakest transport. Thanks, Delton p.s. Would reply to original thread about scramblesuit, but I only just joined the mailing list, and cannot figure out how to reply to preexisting messages.

2 1

Re: [tor-relays] New obfsproxy transport: scramblesuit [bridge operators: please upgrade!]
by Matt 13 Feb '14

13 Feb '14

On Thu, 13 Feb 2014 18:10:38 +0000 George Kadianakis <desnacked(a)riseup.net> wrote: > > Hm, the "Could not set up listener" error message usually appears when > something is already listening on the port that obfsproxy wants to > bind to (Twisted throws the error.CannotListenError exception). > I managed to avoid the problem by running `killall obfsproxy` before restarting Tor. This time around I registered server transport 'obfs3' and registered server transport 'scramblesuit'. Thanks!

1 0

Re: [tor-relays] New obfsproxy transport: scramblesuit [bridge operators: please upgrade!]
by Matt 13 Feb '14

13 Feb '14

On Wed, 12 Feb 2014 18:29:26 +0000 Matt <matt(a)pagan.io> wrote: > I installed obfsproxy through pip. Running Tor 0.2.5.1, I'm seeing > this message: > > [warn] Managed proxy at '/usr/local/bin/obfsproxy' failed the > configuration protocol and will be destroyed. > > That doesn't sound too great. What does this message mean? > Here is more from the debug log: Feb 12 16:38:27.000 [debug] onion_extend_cpath(): Path is complete: 3 steps long Feb 12 16:38:27.000 [debug] circuit_handle_first_hop(): Looking for firsthop '[redacted IPv4]:443' Feb 12 16:38:27.000 [debug] circuit_handle_first_hop(): Conn open. Delivering first onion skin. Feb 12 16:38:27.000 [debug] circuit_send_next_onion_skin(): First skin; sending create cell. Feb 12 16:38:27.000 [debug] circuit_get_by_circid_channel_impl(): circuit_get_by_circid_channel_impl() found nothing for circ_id 2147489864, channel ID 2 (0x2ba73fe349b0) Feb 12 16:38:27.000 [debug] circuit_deliver_create_cell(): Chosen circID 2147489864. Feb 12 16:38:27.000 [debug] circuitmux_attach_circuit(): Attaching circuit 2147489864 on channel 2 to cmux 0x2ba73f9a1c90 Feb 12 16:38:27.000 [debug] circuitmux_attach_circuit(): Attaching circuit 2147489864 on channel 2 to cmux 0x2ba73f9a1c90 Feb 12 16:38:27.000 [debug] append_cell_to_circuit_queue(): Made a circuit active. Feb 12 16:38:27.000 [info] circuit_send_next_onion_skin(): First hop: finished sending CREATE cell to '$5A0EF08FC99F236C5E48FB52E649F092088D009C=ganooloonix at [redacted IPv4]' Feb 12 16:38:27.000 [debug] pt_configure_remaining_proxies(): Configuring remaining managed proxies (1)! Feb 12 16:38:27.000 [info] handle_proxy_line(): Got a line from managed proxy '/usr/local/bin/obfsproxy': (VERSION 1) Feb 12 16:38:27.000 [info] handle_proxy_line(): Got a line from managed proxy '/usr/local/bin/obfsproxy': (SMETHOD-ERROR obfs3 Could not set up listener.) Feb 12 16:38:27.000 [warn] Server managed proxy encountered a method error. (obfs3 Could not set up listener.) Feb 12 16:38:27.000 [warn] Managed proxy at '/usr/local/bin/obfsproxy' failed the configuration protocol and will be destroyed. Feb 12 16:38:27.000 [info] tor_process_handle_destroy(): Terminated process with PID '30355'. Feb 12 16:38:27.000 [info] mark_my_descriptor_dirty(): Decided to publish new relay descriptor: configured managed proxies Feb 12 16:38:27.000 [debug] conn_write_callback(): socket 17 wants to write. Feb 12 16:38:27.000 [debug] flush_chunk_tls(): flushed 1542 bytes, 0 ready to flush, 0 remain. Feb 12 16:38:27.000 [debug] connection_handle_write_impl(): After TLS write of 1542: 0 read, 1589 written Feb 12 16:38:27.000 [debug] channel_write_packed_cell(): Writing packed_cell_t 0x2ba7403cb698 to channel 0x2ba73fe349b0 with global ID 2 Feb 12 16:38:27.000 [debug] channel_flush_from_first_active_circuit(): Made a circuit inactive.

2 1