- tor-project - lists.torproject.org

Future of media and archive.tpo
by Peter Palfrader 28 Mar '16

28 Mar '16

Hey you in the To: line! You are a member of the torarchive or the tormedia unix group. Yet https://trac.torproject.org/projects/tor/wiki/org/operations/Infrastructure lists no maintainer for either archive.torproject.org or media.torproject.org. We will soon have to change the master host for both of these from aroides to some place else. This needs a person or persons who is responsible for the service. If this turns out to be nobody, this will probably mean we have to shut down these services. Please let me know if you think you are the owner of that service. Cheers, -- | .''`. ** Debian ** Peter Palfrader | : :' : The universal https://www.palfrader.org/ | `. `' Operating System | `- https://www.debian.org/

2 1

Summary of meek's costs, February 2016
by David Fifield 28 Mar '16

28 Mar '16

Here's the summary of meek's CDN fees for February 2016. App Engine + Amazon + Azure = total by period all 2014 $600.63 + $917.89 + $0.00 = $1518.52 January 2015 $464.37 + $669.02 + $0.00 = $1133.39 February 2015 $650.53 + $604.83 + $0.00 = $1255.36 March 2015 $690.29 + $815.68 + $0.00 = $1505.97 April 2015 $886.43 + $785.37 + $0.00 = $1671.80 May 2015 $871.64 + $896.39 + $0.00 = $1768.03 June 2015 $601.83 + $820.00 + $0.00 = $1421.83 July 2015 $732.01 + $837.08 + $0.00 = $1569.09 August 2015 $656.76 + $819.59 + $154.89 = $1631.24 September 2015 $617.08 + $710.75 + $490.58 = $1818.41 October 2015 $672.01 + $110.72 + $300.64 = $1083.37 November 2015 $602.35 + $474.13 + $174.18 = $1250.66 December 2015 $561.29 + $603.27 + $172.60 = $1337.16 January 2016 $771.17 + $1581.88 + $329.10 = $2682.15 February 2016 $986.39 + $977.85 + $445.83 = $2410.07 -- total by CDN $10364.78 + $11624.45 + $2067.82 = $24057.05 grand total https://metrics.torproject.org/userstats-bridge-transport.html?graph=userst… In February 2016, meek reached its highest heights ever in terms of number of users--15,000 concurrent--before settling back down to about 10,000, where it started the month. For a few days, when obfs3 was on the way down and obfs4 was on the way up, meek was the #1 most used pluggable transport. obfs4 is a story of its own: it exploded from 3,000 to 15,000 users, probably aided by the many new default high-bandwidth bridges added to Tor Browser recently. It's possible that the increased capacity of obfs4 has taken some of the pressure off meek. We moved over 13 TB of traffic over all three services combined. Reminder that if you configure your own CDN or App Engine account and point it at https://meek.bamsoftware.com/, you will be using a completely unlimited bridge, which will go faster and help save money. == App Engine a.k.a. meek-google == Here is how the Google costs broke down: 5829 GB $699.50 5737 instance hours $286.89 Compared to the previous month: 4274 GB $512.86 5166 instance hours $258.31 https://globe.torproject.org/#/bridge/88F745840F47CE0C6A4FE61D827950B06F9E4… == Amazon a.k.a. meek-amazon == We didn't have the unusually high charge for HTTPS requests this month that we had last month. December had 142M requests in the EU (Ireland) region; January had 805M; and February is back to 156M. I set up additional pricing alarms. Apparently I crossed some kind of threshold for number of alarms, because they started charging for them, for the first time. The alarms cost $0.50. Asia Pacific (Singapore) 247M requests $296.50 952 GB $141.52 Asia Pacific (Sydney) 1M requests $1.95 6 GB $0.91 Asia Pacific (Tokyo) 29M requests $35.42 225 GB $29.68 EU (Ireland) 156M requests $188.05 1222 GB $96.31 South America (Sao Paulo) 9M requests $20.47 77 GB $18.33 US East (Northern Virginia) 94M requests $94.06 687 GB $54.13 -- total 538M requests $636.45 3171 GB $340.88 charge for alarms $0.50 https://globe.torproject.org/#/bridge/F4AD82B2032EDEF6C02C5A529C42CFAFE5165… == Azure a.k.a. meek-azure == Zone 1 4103 GB $357.00 Zone 2 644 GB $ 88.83 -- total 4747 GB $445.83 https://globe.torproject.org/#/bridge/AA033EEB61601B2B7312D89B62AAA23DC3ED8… Earlier reports in this series: https://trac.torproject.org/projects/tor/wiki/doc/meek#Costs

2 2

Re: [tor-project] BWAuth Relay Count Declining
by starlight.2016q1＠binnacle.cx 27 Mar '16

27 Mar '16

At 16:58 3/27/2016 +0000, nusenu wrote: >Hi, > >> For the past several days the number of relays >> measured by my bwuath (whose data is given my >> maatuska) have been slowly declining. Originally >> I had an alert when it was below 8800, a few days >> I noticed it going below that. By today it's >> decreased to below 8700 to 8664. >> >> It's not the lowest bwauth measuring, but I don't >> see anything obviously wrong with my bwauth to >> indicate the decline. (The number I mentioned is >> number of lines in the bwscan.V3BandwidthsFile >> file, not the measured relays that make it into >> the consensus - but the comparison I just made did >> use the consensus count.) > > >If "past several days" matches with "since >2016-03-24", then I would not be surprised if the >Global Frag Networks group [1] caused this - at >least partially. > >Many of the unmeasured relays are from the global >frag group (since they just joined the network) > >https://github.com/ornetstats/stats#stats-on-measured-and-recommended-flag > >Many new relays means a rise in unmeasured relays >and if at the same time a few relays leave then >that would explain the decline even for absolute >values. > > >[1] http://article.gmane.org/gmane.network.onion-routing.ornetradar/1135 > http://article.gmane.org/gmane.network.onion-routing.ornetradar/1132 Interesting. Perhaps the arrival of bunches of new relays is a trigger for or exacerbates the bug where BWauths skip measurements, scanner/band #1 esp.

1 0

Re: [tor-project] BWAuth Relay Count Declining
by starlight.2016q1＠binnacle.cx 27 Mar '16

27 Mar '16

At 09:12 3/27/2016 -0500, you wrote: >For the past several days the number of relays >measured by my bwuath ... have been slowly declining. . . I've noticed the problem as one of my relays has been radically overrated for eight days and is clearly not being measured. The relay currently has consensus of 200% of normal and the link limit is saturated at all times. I might have to shut it down for a couple of days to avoid going over the ISP traffic allocation. Yesterday Faravahar measured it at close to zero (typical after overrating), but the four other BWauths haven't budged. I still plan on launching a project to run a passive BWauth and to fix bugs, eventually restructuring the scanning scheme. But life has not cooperated and I haven't been able to make time for it yet. With luck I should be able to get it underway by May.

1 0

BWAuth Relay Count Declining
by Tom Ritter 27 Mar '16

27 Mar '16

For the past several days the number of relays measured by my bwuath (whose data is given my maatuska) have been slowly declining. Originally I had an alert when it was below 8800, a few days I noticed it going below that. By today it's decreased to below 8700 to 8664. It's not the lowest bwauth measuring, but I don't see anything obviously wrong with my bwauth to indicate the decline. (The number I mentioned is number of lines in the bwscan.V3BandwidthsFile file, not the measured relays that make it into the consensus - but the comparison I just made did use the consensus count.) I wanted to send this email as a potential advance notice; but also ask if there had been any discussion of the bwauth system in Valencia, and if there is any current planned improvements in it. -tom

1 0

Moar Project Ideas!!
by Damian Johnson 25 Mar '16

25 Mar '16

Hopefully the exclamation marks work. Thanks Sebastian for kicking us off! We have just twelve project ideas at present and many of those will go poof due to being outdated. We need moar!!! Please, please give thought to if there's something you'd care to mentor this summer. This is a great opportunity to kickstart development and get a new community member at the same time! We're a bit behind the eight ball right now since Google is evaluating our proposal as we speak, and if selected students will be checking us out on Monday (2/29). Got something? Then just shoot us text like the following and Sebastian or I would be happy to add it... https://www.torproject.org/getinvolved/volunteer.html.en#Coding Cheers! -Damian

12 23

TV producer seeks ordinary Tor use case
by Kate Krauss 24 Mar '16

24 Mar '16

Hi all, I'm working with a TV producer who is doing a three-hour film about the history of the Internet and wants to Interview someone in the US who uses Tor for a sort of every-day use case unrelated to the information security field (person avoiding stalking? Looking at your business competitor? I have a few people but she's not that interested in them. Can you think of anyone who would be good? It seems like a promising film. I'm planning to refer them to LFP but would like to see what I can do to identify other people as well. Cheers, Katie -- Kate Krauss Director of Communications and Public Policy kate(a)torproject.org @TorProject 1-718-864-6647 (works for Signal also) PGP: CC0D 9B42 DE89 D4D0 619B A606 DDEB 3937 7D18 973B -- Kate Krauss Director of Communications and Public Policy kate(a)torproject.org @TorProject 1-718-864-6647 (works for Signal also) PGP: CC0D 9B42 DE89 D4D0 619B A606 DDEB 3937 7D18 973B

2 2

CloudFlare blocking got more severe in December 2015
by David Fifield 24 Mar '16

24 Mar '16

I re-ran some of our analysis scripts from the "Do You See What I See?" paper to bring the graphs up to the present day. In the paper, our OONI-derived data only went up to August 2015, but now it's update to the present. Take a look at this graph, which shows the rate of Tor blocking over time (it's an updated version of Figure 7): https://people.torproject.org/~dcf/graphs/ooni-tor-blocker-rate-20160323.png The interesting characteristic is CloudFlare: after suddenly decreasing from 3% to 1% in June 2015, and slowly decreasing over the course of months to almost 0%, CloudFlare's blocking rate in the OONI data set suddenly increased to about 4% in December 2015. This is consistent with my own experience, that CloudFlare was pretty easy for a while, but lately it's much harder. These percentages are out of all HTTP requests issued by ooniprobe, and only count cases where the Tor request was blocked but the non-Tor request was not blocked. There could also be other confounding factors that we haven't investigated, like maybe OONI's URL list or probe distribution changed suddenly. If you would like to dig deeper, here is the source code that generates the graph: git clone https://www.bamsoftware.com/git/ooni-tor-blocks.git The findblocks.csv file in the source contains a summary of each HTTP request pair issued by ooniprobe, distilled from the raw reports.

1 0

changes to using sudo on torproject.org systems
by Peter Palfrader 22 Mar '16

22 Mar '16

[If you do not use sudo on torproject.org systems, you need not read this.] Comrades, some of you maintain services on torproject.org hosts. Generally, these services run under their dedicated role user, and you use sudo to switch to these roles. Up until now, you have used the LDAP password to authenticate to sudo. We want to change this. The LDAP password is the one you got sent in encrypted mail when your account was first created on db.torproject.org. You might have (should have) changed that on the web-interface [db]. This password is the one that also allows you to log into the management interface [db-login] there and change for instance your mail forwarding configuration or your jabber password [jabber-announce-mail]. The plan is to have a password dedicated to just sudo. To set it, please go to the user management website [db-login] (pick "Update my info"), and set a new (strong) sudo password for yourself. If you want, you can set a password that works for all the hosts that are managed by torproject-admin (*). Alternatively, or additionally, you can have per-host sudo passwords -- just select the appropriate host in the pull-down box. Once set on the web interface, you will have to confirm the new settings by sending a signed challenge to the mail interface. Please ensure you don't introduce any additional line breaks. Note that setting a sudo password will only enable you to use sudo to configured accounts on configured hosts. Consult the output of "sudo -l" if you don't know what you may do. (If you don't know, chances are you don't need to nor can use sudo.) For now, both the LDAP password and the new sudo password will work to authenticate to sudo. Starting in the second week of April, the LDAP password will no longer be accepted for this purpose. If you have any questions, please ask. Thanks, weasel [db] https://db.torproject.org/ [db-login] https://db.torproject.org/login.html [jabber-announce-mail] https://lists.torproject.org/pipermail/tor-project/2016-February/000064.html -- | .''`. ** Debian ** Peter Palfrader | : :' : The universal https://www.palfrader.org/ | `. `' Operating System | `- https://www.debian.org/

1 0

Re: [tor-project] Tor at LibrePlanet 2016
by Nathan Freitas 22 Mar '16

22 Mar '16

On Tue, Mar 22, 2016, at 12:03 PM, David Goulet wrote: > LibrePlanet 2016 was last weekend and we were a couple of Tor people > there (I > won't list them since I didn't ask them before writing this email). > Please, > "other Tor people", feel free to add anything to this thread about what > happened there if you think it would be relevant. I am sorry I could not make it this year, as I was out of town on a family trip. Thank you so much for this report. > I would like to share one thing with all of you that I learned there. It+ > might > not be news to some of you but one part of it below was for me. Through the Guardian Project, we have had numerous interaction with activists, journalists, trainers and organizations based in or supporting communities in Mexico. The triple threat of corporations, government and cartels is real and dire, leaving pretty much no one, no rule of law, or process to trust. Often these various groups are infiltrated by one another, so you end up with mobile operators metadata or surveillance data being acquired by cartels through employees who sell it to them or are pressured into it. With all of this, I can understand how free software and services and projects like Tor must seem like real, tangible, critical allies, and am glad you could connect with them on this topic. > Anyway, I'll stop here. It was a great conference imo and I encourage > anyone > that can attend to do so next year. I'll be there! +n

1 0