wiki:org/teams/MetricsTeam

Metrics Team

Overview

Welcome to the Metrics Team page. The Metrics Team is a group of Tor people who care about measuring and analyzing things in the public Tor network. If you're interested in measurements and analyses, the Metrics Team is what you're looking for. Welcome!

People

The Metrics Team currently consists of between half a dozen and a dozen developers, depending on how and when you count. We're not listing names here to keep the team open to everyone. You're on the team if you're participating in discussions and development, and you're not part of the team anymore if you decide you want to move on (which we hope won't happen).

Communication

Our synchronous medium of communication is IRC meetings that happen every Thursday at 1500 UTC in #tor-meeting on OFTC.

If you want to reach someone from the team between these meetings to ask a development-related question, just go to #tor-dev and mention the magic word "metrics", and somebody from the team might either be around or appear later and get back to you.

Our asynchronous medium of communication is the metrics-team@ mailing list. This list is public in the sense that anyone can subscribe and read archives. But it's moderated on first post, meaning that your first post will be reviewed to make sure it's not spam and on topic and all further posts will go directly to the list. Feel free to subscribe and just listen if you want, and feel free to post if you have a question that you think is on topic.

Products

The Metrics Team is not selling any products, don't worry.

Product is a codebase of software maintained by the Team. Not all products of the Metrics Team are mentioned in this section (for a more complete list see roadmap draft). Instead, we list some long-term products here that are or will be released in way that a third party can decide to run a mirror of this type of service.

Here some links to these products' pages (in alphabetical order):

Releases

Ticket Summary Component Modified
No tickets found

Documentation

An FAQ by and about the Metrics Team. If something you want to know is not answered yet, check back in a few days. Or address your question directly to us via the communication channels listed in the previous section.

The 32C3 State of the Onion talk covered work of the Metrics Team by giving example of how bad relays are detected, investigated, and excluded from the Tor network. The entire talk is worth watching, but if you're short on time, be sure to watch minutes 20 to 28.

More documentation

Censorship Watch

MetricsTimeline lists incidents that might help in interpreting metrics graphs.

Here's a table with tickets related to censorship incidents:

Results (1 - 7 of 16)

1 2 3
Ticket Summary Status Priority Severity Reporter Modified
#34258 Create a shim that hooks up wolpertinger with bridgestrap assigned Medium Normal phw 10 days ago
#34153 Use emma to learn where our bridge distribution mechanisms (don't) work new Medium Normal phw 3 weeks ago
#20348 Allot Communications blocking of vanilla Tor, obfs4, and meek in Kazakhstan, starting 2016-06 reopened Medium Normal dcf 6 weeks ago
#33219 Tor in China (Android) stops at 5% needs_information Medium Normal TiC 2 months ago
#30872 Test BridgeDB's distribution channels in controlled experiment new Medium Normal phw 3 months ago
#30500 Can the GFW still do DPI for "new" vanilla Tor? new Low Normal phw 3 months ago
#31586 Browser problems new Medium Normal fantasy_man59@… 3 months ago
1 2 3

ROADMAP 2020

(Old roadmaps are archived here.)

PRIORITIES 2020

  • OnionPerf: Run the long-running instances, including making major code improvements for Scalability project next year.
  • CollecTor service.
  • Tor Metrics website: We keep most of the Tor Metrics website, including its current graphs.
  • Exit scanner (TorDNSEL rewrite)
  • Running and mantaining Onionoo and Exonerator and Relay Search.

TIMELINE 2020

Component: Internal Services/Tor Sysadmin Team (1 match)

Ticket Summary Component Status Owner Priority Severity Points Reporter
#33733 How do home directories work? Internal Services/Tor Sysadmin Team assigned irl Medium Normal 0.5 irl

Component: Metrics (2 matches)

Ticket Summary Component Status Owner Priority Severity Points Reporter
#24041 Unify Metrics' products operational configuration Metrics assigned metrics-team Medium Normal 2 karsten
#28327 Make sure that each service has at least two operators Metrics new metrics-team High Normal karsten

Component: Metrics/Analysis (2 matches)

Ticket Summary Component Status Owner Priority Severity Points Reporter
#33076 Graph onionperf and consensus information from Rob's experiments Metrics/Analysis needs_review metrics-team Medium Normal 6 mikeperry
#33273 Prop 313: 8.2. Analyse and Monitor IPv6 Stats Metrics/Analysis new metrics-team Medium Normal 3 teor

Component: Metrics/Cloud (12 matches)

Ticket Summary Component Status Owner Priority Severity Points Reporter
#31947 Nagios templates using metrics-team contact Metrics/Cloud new metrics-team Medium Normal 0.5 irl
#33297 Tune Onionperf Nagios alerts Metrics/Cloud new metrics-team Medium Normal 0.5 acute
#33312 Document how to add additional hosts to Nagios Metrics/Cloud new metrics-team Medium Normal 0.5 acute
#33329 Set Lets Encrypt hostname dynamically using ansible facts Metrics/Cloud new metrics-team Medium Normal 0.5 irl
#33394 Automatically build Onionperf documentation from Git Metrics/Cloud new metrics-team Medium Normal 1 acute
#33395 Add option to replace the client and server torrc files Metrics/Cloud new metrics-team Medium Normal 1 acute
#33715 Create a metrics-common role and ops doc Metrics/Cloud accepted irl Medium Normal 3.4 irl
#33717 Define metrics-common group vars to replace exit-scanner-sys role Metrics/Cloud accepted irl Medium Normal 0.2 irl
#33718 Check check.torproject.org in Nagios Metrics/Cloud new metrics-team Medium Normal 0.2 irl
#33719 Check DNSEL in Nagios Metrics/Cloud new metrics-team Medium Normal 0.2 irl
#33720 Check exit scanner in Nagios Metrics/Cloud new metrics-team Medium Normal 0.2 irl
#31950 Use longer timeout for op-hk pings Metrics/Cloud new metrics-team High Normal 0.2 irl

Component: Metrics/Exit Scanner (1 match)

Ticket Summary Component Status Owner Priority Severity Points Reporter
#33508 Write Ops Doc for check service Metrics/Exit Scanner accepted irl Medium Normal 1.2 irl

Component: Metrics/Onionperf (29 matches)

Ticket Summary Component Status Owner Priority Severity Points Reporter
#29370 Measure mode with arbitrary tgen traffic models Metrics/Onionperf reopened metrics-team Low Normal 1 irl
#28271 Check OnionPerf instances from Nagios Metrics/Onionperf assigned metrics-team Medium Normal 1 irl
#29365 Add digests and sizes to index.xml Metrics/Onionperf reopened metrics-team Medium Normal 0.5 irl
#29368 Add single onion service mode Metrics/Onionperf needs_revision phw Medium Normal 2 irl
#29369 Fix message logging and filtering Metrics/Onionperf needs_review phw Medium Normal 1.0 irl
#30362 OnionPerf TGen parser needs reworking Metrics/Onionperf reopened metrics-team Medium Normal 3 acute
#33257 Add CDF-DL graph Metrics/Onionperf needs_review karsten Medium Normal 3 karsten
#33259 Store measurements in a local database to reduce plotting time Metrics/Onionperf new metrics-team Medium Normal 5 karsten
#33260 Add option to filter graphed OnionPerf results by relay fingerprint Metrics/Onionperf new metrics-team Medium Normal 3 karsten
#33318 O1.1 Improve monitoring: We will produce a Nagios plugin for monitoring OnionPerf instances to ensure that they are operating correctly. Metrics/Onionperf new metrics-team Medium Normal gaba
#33319 O1.2 Improve ease of deployment and maintenance: We will produce Ansible tasks for deploying and managing deployments of OnionPerf instances, which also allow for performing upgrades and custom configuration changes. Metrics/Onionperf new metrics-team Medium Normal gaba
#33320 Objective 1: Make operational improvements to existing OnionPerf deployments and make it easier to deploy new OnionPerf instances Metrics/Onionperf new metrics-team Medium Normal gaba
#33321 O1.3 Make OnionPerf more accessible to researchers and developers Metrics/Onionperf new metrics-team Medium Normal gaba
#33323 O2.1 Add instance metadata: We need a way to distinguish our current four long-term OnionPerf measurements that are automatically published to the Metrics portal from short-term experimental measurements. Metrics/Onionperf new metrics-team Medium Normal gaba
#33324 O2.2 Develop at least one new OnionPerf model: An OnionPerf model defines the pattern of traffic that is generated by the instances for conducting a measurement. Metrics/Onionperf new metrics-team Medium Normal gaba
#33325 O2.3 Implement static guard node support for OnionPerf. Metrics/Onionperf new metrics-team Medium Normal gaba
#33326 Objective 3: Make improvements to the way we analyze performance metrics. Metrics/Onionperf new metrics-team Medium Normal gaba
#33327 O3.1 Develop developer-facing tooling to quickly graph baseline performance metrics. Metrics/Onionperf new metrics-team Medium Normal gaba
#33328 O3.2 Include additional OnionPerf filters. Metrics/Onionperf new metrics-team Medium Normal gaba
#33391 Add new metadata fields and definitions Metrics/Onionperf needs_review metrics-team Medium Normal 1 acute
#33392 Add new metadata fields to json output Metrics/Onionperf new metrics-team Medium Normal 0.5 acute
#33399 Measure static guard nodes with OnionPerf Metrics/Onionperf new metrics-team Medium Normal 4 acute
#33419 Add analysis to support static guard measurements Metrics/Onionperf new metrics-team Medium Normal 2 acute
#33420 Add CBT events to Onionperf result files Metrics/Onionperf new metrics-team Medium Normal 1 acute
#33421 Track which Guard is used for experimental measurements Metrics/Onionperf new metrics-team Medium Normal 3 acute
#33422 Include more events about CBT in results Metrics/Onionperf new metrics-team Medium Normal 1 acute
#33438 OnionPerf: Scalability, Performance, Establishing Baseline Metrics Metrics/Onionperf new metrics-team Medium Normal gaba
#33974 Update OnionPerf to TGen 1.0.0 Metrics/Onionperf assigned jnewsome Medium Normal 2 karsten
#30798 Develop and deploy tgen model resembling ping Metrics/Onionperf reopened metrics-team High Normal 5 karsten

Component: Metrics/Website (1 match)

Ticket Summary Component Status Owner Priority Severity Points Reporter
#33397 Update metrics-web to only plot "official" data Metrics/Website new metrics-team Medium Normal 0.5 acute

ALL Open Tickets

The following table lists all open tickets, from recently modified to not touched in years, of products that are maintained by Metrics Team people or otherwise related. Not all of these products are actively maintained, so be sure to reach out to the team before starting to hack on any of these and also take a look at the Volunteer's Guide. Also note that not all metrics-related tools use Tor's Trac as their bug tracker, so you might also want to ask for bug trackers of other metrics tools.

Results (1 - 20 of 333)

1 2 3 4 5 6 7 8 9 10 11
Ticket Summary Component Status Owner Priority Severity Created Modified Reporter
#34257 Analyze unusual distribution of time to extend to first hop in circuit Metrics/Onionperf new metrics-team Medium Normal 10 days ago 2 hours ago karsten
#29368 Add single onion service mode Metrics/Onionperf needs_revision phw Medium Normal 16 months ago 14 hours ago irl
#29369 Fix message logging and filtering Metrics/Onionperf needs_review phw Medium Normal 16 months ago 15 hours ago irl
#34261 Remove pandas warning when using bullseye version of pandas Metrics/Onionperf needs_review karsten Medium Normal 10 days ago 45 hours ago karsten
#33257 Add CDF-DL graph Metrics/Onionperf needs_review karsten Medium Normal 4 months ago 2 days ago karsten
#34231 Document and maybe improve how we're mapping TGen transfers to Tor streams/circuits Metrics/Onionperf new metrics-team Medium Normal 2 weeks ago 3 days ago karsten
#34316 Make -o/-i arguments mutually exclusive Metrics/Onionperf needs_review karsten Medium Normal 4 days ago 4 days ago karsten
#33432 Multiple downloads for oneshot mode Metrics/Onionperf needs_information metrics-team Medium Normal 3 months ago 4 days ago acute
#33111 upgrade all metrics hosts to Debian buster Metrics new metrics-team Medium Normal 4 months ago 6 days ago anarcat
#33010 Monitor cloudflare captcha rate: do a periodic onionperf-like query to a cloudflare-hosted static site Metrics/Ideas new metrics-team Medium Normal 4 months ago 6 days ago arma
#34191 Combine multiple analysis files into single data set Metrics/Onionperf accepted acute Medium Normal 3 weeks ago 6 days ago karsten
#34297 Explore different options other than crontab to have a flexible scheduling system Metrics/Ideas new metrics-team Low Normal 8 days ago 8 days ago woswos
#34293 Create an API for running the system on the user-provided websites Metrics/Ideas new metrics-team Very Low Normal 9 days ago 8 days ago woswos
#34292 Create an API for people to fetch data easily Metrics/Ideas new metrics-team Very Low Normal 9 days ago 8 days ago woswos
#34291 Create an API for enabling 3rd party interactions with the system Metrics/Ideas new metrics-team Very Low Normal 9 days ago 8 days ago woswos
#34290 Enhance the available visualizations on the dashboard Metrics/Ideas new metrics-team Low Normal 9 days ago 8 days ago woswos
#34289 Integrate Cloudflare API not to change Cloudflare settings manually Metrics/Ideas new metrics-team Medium Normal 9 days ago 8 days ago woswos
#34296 Develop a mechanism for disabling/enabling cookie, JavaScript, etc. functionality of the web browsers Metrics/Ideas new metrics-team Medium Normal 8 days ago 8 days ago woswos
#34295 Develop a module for parsing and modifying the HTTP headers Metrics/Ideas new metrics-team Medium Normal 8 days ago 8 days ago woswos
#34288 Integrate more web browsers/fetchers and their older versions Metrics/Ideas new metrics-team Medium Normal 9 days ago 8 days ago woswos
1 2 3 4 5 6 7 8 9 10 11

Last modified 4 weeks ago Last modified on May 4, 2020, 4:13:36 PM

Attachments (12)

Download all attachments as: .zip