tor-commits July 2020

tor-commits@lists.torproject.org

17 participants
2100 discussions

[onionperf/master] Make README.md more useful.
by karsten＠torproject.org 02 Jul '20

02 Jul '20

commit 54e0bd21227ddb99b64a7d8eb8ca0c0457936d92 Author: Karsten Loesing <karsten.loesing(a)gmx.net> Date: Tue Jun 30 17:55:18 2020 +0200 Make README.md more useful. This commit attempts to make README.md the primary, comprehensive documentation for OnionPerf. There will be more documentation on the wiki for things that change faster than we want to update a file in Git or that are too specific for most users. But this README.md is supposed to cover 90% of the questions that new users would have. Implements #40001. --- README.md | 532 +++++++++++++++++++++----------------------------------------- 1 file changed, 181 insertions(+), 351 deletions(-) diff --git a/README.md b/README.md index 35d3b7d..9f0faf4 100644 --- a/README.md +++ b/README.md @@ -1,446 +1,276 @@ # OnionPerf -OnionPerf is a utility to track Tor and onion service performance. - -OnionPerf uses multiple processes and threads to download random data -through Tor while tracking the performance of those downloads. The data is -served and fetched on localhost using two TGen (traffic generator) -processes, and is transferred through Tor using Tor client processes and -an ephemeral Tor Onion Service. Tor control information and TGen -performance statistics are logged to disk, analyzed once per day to -produce a json stats database and files that can feed into Torperf, and -can later be used to visualize changes in Tor client performance over time. - -For more information, see https://git.torproject.org/onionperf - -## Table of contents - -* [Quick deployment instructions](#quick-deployment-instructions) -* [Step-by-step installation instructions](#step-by-step-installation-instructions) - - [Get OnionPerf](#get-onionperf) - - [Install System Dependencies](#install-system-dependencies) - - [Install Python modules](#install-python-modules) - - [Build Tor](#build-tor) - - [Build TGen Traffic Generator](#build-tgen-traffic-generator) - - [Build and Install OnionPerf](#build-and-install-onionperf) -* [Run OnionPerf](#run-onionperf) - - [Measure Tor](#measure-tor) - - [Analyze and Visualize Results](#analyze-and-visualize-results) - - [Troubleshooting](#troubleshooting) -* [Contribute](#contribute) - -## Quick deployment instructions - -These are the quick deployment instructions for the current Debian stable distribution. +[TOC] -``` -sudo apt install git cmake make build-essential gcc libigraph0-dev libglib2.0-dev python3-dev libxml2-dev python3-lxml python3-networkx python3-scipy python3-matplotlib python3-numpy libevent-dev libssl-dev python3-stem python3-pandas python3-seaborn python3-setuptools tor +## Overview -git clone https://github.com/shadow/tgen.git -cd tgen -mkdir build -cd build -cmake .. -DCMAKE_INSTALL_PREFIX=/home/$USER/.local -make -sudo ln -s ~/tgen/build/tgen /usr/bin/tgen +### What does OnionPerf do? -git clone https://github.com/torproject/onionperf -cd onionperf -python3 setup.py build -python3 setup.py install --user -``` +OnionPerf measures performance of bulk file downloads over Tor. Together with its predecessor, Torperf, Onionperf has been used to measure long-term performance trends in the Tor network since 2009. It is also being used to perform short-term performance experiments to compare different Tor configurations or implementations. -Once the installation finished, you can find the onionperf script in -`~/.local/bin/`. +OnionPerf uses multiple processes and threads to download random data through Tor while tracking the performance of those downloads. The data is served and fetched on localhost using two TGen (traffic generator) processes, and is transferred through Tor using Tor client processes and an ephemeral Tor onion service. Tor control information and TGen performance statistics are logged to disk and analyzed once per day to produce a JSON analysis file that can later be used to visualize changes in Tor client performance over time. -## Step-by-step installation instructions +### What does OnionPerf *not* do? -Here you can find more detailed instructions for the current Debian stable distribution. +OnionPerf does not attempt to simulate complex traffic patterns like a web-browsing user or a voice-chatting user. It measures a very specific user model: a bulk 5 MiB file download over Tor. -### Get OnionPerf +OnionPerf does not interfere with how Tor selects paths and builds circuits, other than setting configuration values as specified by the user. As a result it cannot be used to measure specific relays nor to scan the entire Tor network. -``` -git clone https://git.torproject.org/onionperf.git -cd onionperf -``` +## Installation -### Install System Dependencies +OnionPerf has a couple dependencies in order to perform measurements or analyze and visualize measurement results. These dependencies include Tor, TGen (traffic generator), and a couple Python packages. - + **Tor** (>= v0.2.7.3-rc): libevent, openssl - + **TGen** (Shadow >= v1.11.1): cmake, glib2, igraph - + **OnionPerf**: python3 +The following description was written with a Debian system in mind but should be transferable to other Linux distributions and possibly even other operating systems. -The easiest way to satisfy all system dependencies is to use a package manager. TGen is not currently packaged and needs to be built from source. -Note we only provide support for the current Debian Stable distribution. +### Tor -``` -sudo apt install cmake make build-essential gcc libigraph0-dev libglib2.0-dev python3-dev +OnionPerf relies on the `tor` binary to start a Tor client on the client side and another one on the server side to host onion services. + +The easiest way is to install the `tor` package, which puts the `tor` binary into the `PATH` where OnionPerf will find it. Optionally, systemd can be instructed to make sure that `tor` is never started as a service: + +```shell +sudo apt install tor +sudo systemctl stop tor.service +sudo systemctl mask tor.service ``` -### Install Python modules +Alternatively, Tor can be built from source: - + **OnionPerf** python modules: stem (>= v1.7.0), lxml, networkx, numpy, matplotlib. +```shell +sudo apt install automake build-essential libevent-dev libssl-dev zlib1g-dev +cd ~/ +git clone https://git.torproject.org/tor.git +cd tor/ +./autogen.sh +./configure --disable-asciidoc +make +``` -#### Option 1: Package Manager +In this case the resulting `tor` binary can be found in `~/tor/src/app/tor` and needs to be passed to OnionPerf's `--tor` parameter when doing measurements. -The easiest way to satisfy all system dependencies is to use a package manager. +### TGen -``` -apt install tor libxml2-dev python3-lxml python3-networkx python3-scipy python3-matplotlib python3-numpy python3-stem python3-pandas python3-seaborn python3-setuptools +OnionPerf uses TGen to generate traffic on client and server side for its measurements. Installing dependencies, cloning TGen to a subdirectory in the user's home directory, checking out version 0.0.1, and building TGen is done as follows: +```shell +sudo apt install cmake libglib2.0-dev libigraph0-dev make +cd ~/ +git clone https://github.com/shadow/tgen.git +cd tgen/ +git checkout -b v0.0.1 v0.0.1 +mkdir build +cd build/ +cmake .. +make ``` -#### Option 2: pip +The TGen binary will be contained in `~/tgen/build/tgen`, which is also the path that needs to be passed to OnionPerf's `--tgen` parameter when doing measurements. -Python modules can also be installed using `pip`. The python modules that are -required for each OnionPerf subcommand are as follows: +### OnionPerf - + `onionperf monitor`: stem - + `onionperf measure`: stem, lxml, networkx - + `onionperf analyze`: stem - + `onionperf visualize`: scipy, numpy, pylab, matplotlib +OnionPerf is written in Python 3. The following instructions assume that a Python virtual environment is being used, even though installation is also possible without that. -You must first satisfy the system/library requirements of each of the python modules. -Note: pip installation is not recommended as software installed by pip is not verified. +The virtual environment is being created, activated, and tested using: +```shell +sudo apt install python3-venv +cd ~/ +python3 -m venv venv +source venv/bin/activate +which python3 ``` -sudo apt-get install python3-dev libxml2 libxml2-dev libxslt1.1 libxslt1-dev libpng12-0 libpng12-dev libfreetype6 libfreetype6-dev + +The last command should output something like `~/venv/bin/python3` as the path to the `python3` binary used in the virtual environment. + +In the next step, the OnionPerf repository is being cloned and requirements are being installed: + +```shell +git clone https://git.torproject.org/onionperf.git +pip3 install --no-cache -r onionperf/requirements.txt ``` -It is recommended to use virtual environments to keep all of the dependencies self-contained and -to avoid conflicts with your other python projects. +The next step after that is to install OnionPerf and print out the usage information to see if the installation was successful: +```shell +cd onionperf/ +python3 setup.py install +cd ~/ +onionperf --help ``` -python3 -m venv venv -source venv/bin/activate -pip install --upgrade pip # make sure that you have a recent version of pip -pip install -r requirements.txt # installs all required python modules for all OnionPerf subcommands + +The virtual environment is deactivated with the following command: + +```shell deactivate ``` -If you don't want to use virtualenv, you can install with: +However, in order to perform measurements or doing analyses, the virtual environment needs to be activated first. This will ensure all the paths are found. -``` -pip install stem lxml networkx scipy numpy matplotlib -``` -**Note**: You may want to skip installing numpy and matplotlib if you don't -plan to use the `visualize` subcommand, since those tend to require several -large dependencies. +## Measurement -### Build Tor +Performing measurements with OnionPerf is done by starting an `onionperf` process that itself starts several other processes and keeps running until it is interrupted by the user. During this time it performs new measurements every 5 minutes and logs measurement results to files. -**Note**: You can install Tor with apt, although the -preferred method is to build from source. To install using from backports: +Ideally, OnionPerf is run detached from the terminal session using tmux, systemd, or similar, except for the most simple test runs. The specifics for using these tools are not covered in this document. -``` -echo 'deb http://deb.debian.org/debian stretch-backports main' >> /etc/apt/sources.list -apt update -apt-get -t stretch-backports install tor -``` -Or, if building from source: +### Starting and stopping measurements -``` -apt install libevent libevent-dev libssl-dev -``` +The most trivial configuration is to measure onion services only. In that case, OnionPerf is taking care of all configurations, and there are no firewall rules or port forwards to take care of. -``` -git clone https://git.torproject.org/tor.git -cd tor -./autogen.sh -./configure --disable-asciidoc -make +Starting these measurements is as simple as: + +```shell +cd ~/ +onionperf measure --onion-only --tgen ~/tgen/build/tgen --tor ~/tor/src/app/tor ``` -### Build TGen Traffic Generator +OnionPerf logs its main output on the console and then waits indefinitely until the user presses `CTRL-C` for graceful shutdown. It does not, however, print out measurement results or progress on the console, just a heartbeat message every hour. -The traffic generator currently exists in the Shadow simulator repository, -but we will build TGen as an external tool and skip building both the full -simulator and the TGen simulator plugin. +OnionPerf's `measure` mode has several command-line parameters for customizing measurements. See the following command for usage information: -``` -git clone https://github.com/shadow/tgen.git -cd tgen -mkdir build -cd build -cmake .. -DCMAKE_INSTALL_PREFIX=/home/$USER/.local -make -ln -s build/tgen /usr/bin/tgen +```shell +onionperf measure --help ``` -### Build and Install OnionPerf +### Output directories and files -If using pip and virtualenv (run from onionperf base directory): +OnionPerf writes several files to two subdirectories in the current working directory while doing measurements: -``` -source venv/bin/activate -pip install -I . -deactivate -``` +- `onionperf-data/` is the main directory containing measurement results. + - `htdocs/` is created at the first UTC midnight after starting and contains measurement analysis result files that can be shared via a local web server. + - `$date.onionperf.analysis.json.xz` contains extracted metrics in OnionPerf's analysis JSON format. + - `index.xml` contains a directory index with file names, sizes, last-modified times, and SHA-256 digests. + - `tgen-client/` is the working directory of the client-side `tgen` process. + - `log_archive/` is created at the first UTC midnight after starting and contains compressed log files from previous UTC days. + - `onionperf.tgen.log` is the current log file. + - `tgen.graphml.xml` is the traffic model file generated by OnionPerf and used by TGen. + - `tgen-server/` is the working directory of the server-side `tgen` process with the same structure as `tgen-client/`. + - `tor-client/` is the working directory of the client-side `tor` process. + - `log_archive/` is created at the first UTC midnight after starting and contains compressed log files from previous UTC days. + - `onionperf.tor.log` is the current log file containing log messages by the client-side `tor` process. + - `onionperf.torctl.log` is the current log file containing controller events obtained by OnionPerf connecting to the control port of the client-side `tor` process. + - `[...]` (several other files written by the client-side `tor` process to its data directory) + - `tor-server/` is the working directory of the server-side `tor` process with the same structure as `tor-client/`. +- `onionperf-private/` contains private keys of the onion services used for measurements and potentially other files that are not meant to be published together with measurement results. -If using just pip: +### Changing Tor configurations -``` -pip install -I . -``` +OnionPerf generates Tor configurations for both client-side and server-side `tor` processes. There are a few ways to add Tor configuration lines: -Otherwise: +- If the `BASETORRC` environment variable is set, OnionPerf appends its own configuration options to the contents of that variable. Example: -``` -python3 setup.py build -python3 setup.py install -``` + ```shell + BASETORRC=$'Option1 Foo\nOption2 Bar\n' onionperf ... + ``` -The installation step for Onionperf can be omitted, and instead Onionperf can run directly from the cloned repository: +- If the `--torclient-conf-file` and/or `--torserver-conf-file` command-line arguments are given, the contents of those files are appended to the configurations of client-side and/or server-side `tor` process. +- If the `--additional-client-conf` command-line argument is given, its content is appended to the configuration of the client-side `tor` process. -``` -git clone https://github.com/torproject/onionperf -cd onionperf/onionperf -PYTHONPATH=.. python3 onionperf +These options can be used, for example, to change the default measurement setup use bridges (or pluggable transports) by passing bridge addresses as additional client configuration lines as follows: + +```shell +onionperf measure --additional-client-conf="UseBridges 1\nBridge 72.14.177.231:9001 AC0AD4107545D4AF2A595BC586255DEA70AF119D\nBridge 195.91.239.8:9001 BA83F62551545655BBEBBFF353A45438D73FD45A\nBridge 148.63.111.136:35577 768C8F8313FF9FF8BBC915898343BC8B238F3770" ``` -Dependencies (with the exception of ```python3-setuptools```) must be installed for Onionperf to run in this way. +### Changing the TGen traffic model -## Run OnionPerf +OnionPerf is a relatively simple tool that can be adapted to do more complex measurements beyond what can be configured on the command line. -OnionPerf has several modes of operation and a help menu for each. For a -description of each mode, use: +For example, the hard-coded traffic model generated by OnionPerf and executed by the TGen processes is to send a small request from client to server and receive a relatively large response of 5 MiB of random data back. This model can be changed by editing `~/onionperf/onionperf/model.py`, rebuilding, and restarting measurements. For specifics, see the TGen documentation. -``` -onionperf -h -``` +### Sharing measurement results - + **monitor**: Connect to Tor and log controller events to file - + **measure**: Measure Tor and Onion Service Performance using TGen - + **analyze**: Analyze Tor and TGen output - + **visualize**: Visualize OnionPerf analysis results +Measurement results can be further analyzed and visualized on the measuring host. But in many cases it's more conventient to do analysis and visualization on another host, also to compare measurements from different hosts to each other. -### Measure Tor +There are at least two common ways of sharing measurement results: -To run in measure mode, you will need to give OnionPerf the path to your custom -'tor' and 'tgen' binary files if they do not exist in your PATH -environment variable. +1. Creating a tarball of the `onionperf-data/` directory; and +2. Using a local web server to serve the contents of the `onionperf-data/` directory. -``` -./onionperf measure --tor=/home/rob/tor/src/or/tor \ ---tgen=/home/rob/shadow/src/plugin/shadow-plugin-tgen/build/tgen -``` +The details of doing either of these two methods are not covered in this document. -This will run OnionPerf in measure mode with default ports; a TGen server runs -on port 8080. Port 8080 **must** be open on your firewall if you want to do -performance measurements with downloads that exit the Tor network. - -By default, OnionPerf will run a TGen client/server pair that transfer traffic -through Tor and through an ephemeral onion service started by OnionPerf. TGen and Tor -log data is collected and stored beneath the `onionperf-data` directory, and other -information about Tor's state during the measurement process is collected from Tor's -control port and logged to disk. - -While running, OnionPerf log output is saved to component-specific log files. -Log files for each OnionPerf component (tgen client, tgen server, tor client, tor -server) are stored in their own directory, under `onionperf-data`: - - + `tgen-client/onionperf.tgen.log` - + `tgen-server/onionperf.tgen.log` - + `tor-client/onionperf.torctl.log` - + `tor-client/onionperf.tor.log` - + `tor-server/onionperf.torctl.log` - + `tor-server/onionperf.tor.log` - -Every night at 11:59 UTC, OnionPerf will analyze the latest results from these log -files using the same parsing functions that are used in the `onionperf analyze` -subcommand (which is described in more detail below). The analysis produces a -`onionperf.analysis.json` stats file that contains numerous measurements and other -contextual information collected during the measurement process. The `README_JSON.md` -file in this repo describes the format and elements contained in the `json` file. - -The daily generated `json` files are placed in the web docroot and are -available through the local filesystem in the `onionperf-data/htdocs` directory. -These files can be shared via a web server such as Apache or nginx. - -Once the analysis is complete, OnionPerf will rotate all log files; each log file -is moved to a `log_rotate` subdirectory and renamed to include a timestamp. Each -component has its own collection of log files with timestamps in their own `log_rotate` -subdirectories. - -You can reproduce the same `json` file that is automatically produced every day while -running in `onionperf measure` mode. To do this, you would run `onionperf analyze` on -specific log files from the `log_rotate` directories. You can also plot the measurement -results from the `json` files by running in `onionperf visualize` mode. See below for -more details. - -#### Measure Tor using arbitrary traffic models - -OnionPerf uses `tgen` to serve and fetch data through Tor. By default, the -traffic is generated using a bulk download model: each transfer performed is -5MB in size, and transfers are spaced 5 minutes apart. - -The class `TorPerfModel` defined in `onionperf/model.py` is used to generate -this traffic model at runtime: +### Troubleshooting -``` -class TorperfModel(GeneratableTGenModel): - - def __init__(self, tgen_port="8889", tgen_servers=["127.0.0.1:8888"], socksproxy=None): - self.tgen_port = tgen_port - self.tgen_servers = tgen_servers - self.socksproxy = socksproxy - self.graph = self.generate() - - def generate(self): - server_str = ','.join(self.tgen_servers) - g = DiGraph() - - if self.socksproxy is not None: - g.add_node("start", serverport=self.tgen_port, peers=server_str, loglevel="info", heartbeat="1 minute", socksproxy=self.socksproxy) - else: - g.add_node("start", serverport=self.tgen_port, peers=server_str, loglevel="info", heartbeat="1 minute") - g.add_node("pause", time="5 minutes") - g.add_node("transfer5m", type="get", protocol="tcp", size="5 MiB", timeout="270 seconds", stallout="0 seconds") - g.add_edge("start", "pause") - g.add_edge("pause", "pause") - g.add_edge("pause", "transfer5m") - - return g -``` +If anything goes wrong while doing measurements, OnionPerf typically informs the user in its console output. This is also the first place to look for investigating any issues. -Although OnionPerf does not currently support other TGen traffic models out of -the box, it is possible to modify the code shown above to change the -behaviour of the default model. +The second place would be to check the log files in `~/onionperf-data/tgen-client/` or `~/onionperf-data/tor-client/`. -All graph nodes and edges added after the "start" node can be redefined to model -any custom traffic behaviour. Refer to the [TGen -documentation](https://github.com/shadow/tgen/blob/master/doc/TGen-Overview.md) -and [list of traffic model graph -examples](https://github.com/shadow/tgen/blob/master/tools/scripts/generate_tgen_config.py). -For example, to model web traffic behaviour, the class can be modified as follows: +The most common configuration problems are probably related to firewall and port forwarding for doing direct (non onion-service) measurements. The specifics for setting up the firewall are out of scope for this document. -``` -class TorperfModel(GeneratableTGenModel): - - def __init__(self, tgen_port="8889", tgen_servers=["127.0.0.1:8888"], socksproxy=None): - self.tgen_port = tgen_port - self.tgen_servers = tgen_servers - self.socksproxy = socksproxy - self.graph = self.generate() - - def generate(self): - server_str = ','.join(self.tgen_servers) - g = DiGraph() - - if self.socksproxy is not None: - g.add_node("start", serverport=self.tgen_port, peers=server_str, loglevel="info", heartbeat="1 minute", socksproxy=self.socksproxy) - else: - g.add_node("start", serverport=self.tgen_port, peers=server_str, loglevel="info", heartbeat="1 minute") - - g.add_node("streamA", sendsize="1 kib", recvsize="1 mib") - g.add_node("streamB1", sendsize="10 kib", recvsize="1 mib") - g.add_node("streamB2", sendsize="100 kib", recvsize="10 mib") - g.add_node("streamB3", sendsize="1 mib", recvsize="100 mib") - - g.add_node("pause_sync") - g.add_node("pause", time="1,2,3,4,5,6,7,8,9,10") - g.add_node("end", time="1 minute", count="30", recvsize="1 GiB", sendsize="1 GiB") - - # a small first round request - g.add_edge("start", "streamA") - # second round requests in parallel - g.add_edge("streamA", "streamB1") - g.add_edge("streamA", "streamB2") - g.add_edge("streamA", "streamB3") - # wait for both second round streams to finish - g.add_edge("streamB1", "pause_sync") - g.add_edge("streamB2", "pause_sync") - g.add_edge("streamB3", "pause_sync") - # check if we should stop - g.add_edge("pause_sync", "end") - g.add_edge("end", "pause") - g.add_edge("pause", "start") - - return g -``` +Another class of common issues of long-running measurements is that one of the `tgen` or `tor` processes dies for reasons or hints (hopefully) to be found in their respective log files. -The new traffic model will only be used for measurements once OnionPerf is reinstalled. +In order to avoid extended downtimes it is recommended to deploy monitoring tools that check whether measurement results produced by OnionPerf are fresh. The specifics are, again, out of scope for this document. -Note that custom code may be required to analyze the log files and visualize the data. +## Analysis -### Analyze and Visualize Results +The next steps after performing measurements are to analyze and optionally visualize measurement results. -OnionPerf runs the data it collects through `analyze` mode every night at midnight to -produce the `onionperf.analysis.json` stats file. This file can be reproduced by using -`onionperf analyze` mode and feeding in a TGen log file from the -`onionperf-data/tgen-client/log_archive` directory and the matching Torctl log file from -the `onionperf-data/tor-client/log_archive` directory. +### Analyzing measurement results -For example: +While performing measurements, OnionPerf writes quite verbose log files to disk. The first step in the analysis is to parse these log files, extract key metrics, and write smaller and more structured measurement results to disk. This is done with OnionPerf's `analyze` mode. -``` -onionperf analyze --tgen onionperf-data/tgen-client/log_archive/onionperf_2015-11-15_15\:59\:59.tgen.log \ ---torctl onionperf-data/tor-client/log_archive/onionperf_2015-11-15_15\:59\:59.torctl.log +For example, the following command analyzes current log files of a running (or stopped) OnionPerf instance (as opposed to log-rotated, compressed files from previous days): + +```shell +onionperf analyze --tgen ~/onionperf-data/tgen-client/onionperf.tgen.log --torctl ~/onionperf-data/tor-client/onionperf.torctl.log ``` -This produces the `onionperf.analysis.json` file, which can then be plotted like so: +The output analysis file is written to `onionperf.analysis.json.xz` in the current working directory. The file format is described in more detail in `README_JSON.md`. -``` -onionperf visualize --data onionperf.analysis.json "onionperf-test" -``` +The same analysis files are written automatically as part of ongoing measurements once per day at UTC midnight and can be found in `onionperf-data/htdocs/`. + +OnionPerf's `analyze` mode has several command-line parameters for customizing the analysis step: -This will save new PDFs containing several graphs in the current directory. These include: +```shell +onionperf analyze --help +``` -#### TGen +### Visualizing measurement results -- Number of transfer AUTH errors, each client -- Number of transfer PROXY errors, each client -- Number of transfer AUTH errors, all clients over time -- Number of transfer PROXY errors, all clients over time -- Bytes transferred before AUTH error, all downloads -- Bytes transferred before PROXY error, all downloads -- Median bytes transferred before AUTH error, each client -- Median bytes transferred before PROXY error, each client -- Mean bytes transferred before AUTH error, each client -- Mean bytes transferred before PROXY error, each client +Step two in the analysis is to process analysis files with OnionPerf's `visualize` mode which produces CSV and PDF files as output. -#### Tor +For example, the analysis file produced above can be visualized with the following command, using "Test Measurements" as label for the data set: -- 60 second moving average throughput, read, all relays -- 1 second throughput, read, all relays -- 1 second throughput, read, each relay -- 60 second moving average throughput, write, all relays -- 1 second throughput, write, all relays -- 1 second throughput, write, each relay +```shell +onionperf visualize --data onionperf.analysis.json.xz "Test Measurements" +``` -### Troubleshooting +As a result, two files are written to the current working directory: -While OnionPerf is running, it will log heartbeat status messages to indicate -the health of the subprocesses. It also monitors these subprocesses with -"watchdog threads" and restarts each subprocess if they fail. If more than 10 -failures happen within 60 minutes, then the watchdog will give up and exit, -and at that point the heartbeat message should indicate that one of the -subprocesses died. +- `onionperf.viz.$datetime.csv` contains visualized data in a CSV file format; and +- `onionperf.viz.$datetime.pdf` contains visualizations in a PDF file format. -While running, a number of `Warning` messages might be logged. For example: +Similar to the other modes, OnionPerf's `visualize` mode has command-line parameters for customizing the visualization step: -``` -2016-12-21 12:15:17 1482318917.473774 [onionperf] [WARNING] command -'/home/rob/shadow/src/plugin/shadow-plugin-tgen/build/tgen -/home/rob/onionperf/onionperf-data/tgen-server/tgen.graphml.xml' -finished before expected" +```shell +onionperf visualize --help ``` -This specific warning indicates that the tgen server watchdog thread detected -that the tgen server prematurely exited. The watchdog should have spun -up another tgen server to replace the one that died. +### Interpreting the PDF output format -To find out why this is happening you should check the specific component logs -which are all subdirectories of the onionperf-data directory. -In this particular case tgen-server log file revealed the problem: +The PDF output file contains visualizations of the following metrics: -``` -2016-12-20 18:46:29 1482255989.329712 [critical] [shd-tgen-server.c:94] [tgenserver_new] bind(): socket 5 returned -1 error 98: Address already in use -``` +- Time to download first (last) byte, which is defined as elapsed time between starting a measurement and receiving the first (last) byte of the HTTP response. +- Throughput, which is computed from the elapsed time between receiving 0.5 and 1 MiB of the response. +- Number of downloads. +- Number and type of failures. + +### Interpreting the CSV output format + +The CSV output file contains the same data that is visualized in the PDF file. It contains the following columns: + +- `transfer_id` is the identifier used in the TGen client logs which may be useful to look up more details about a specific measurement. +- `error_code` is an optional error code if a measurement did not succeed. +- `filesize_bytes` is the requested file size in bytes. +- `label` is the data set label as given in the `--data/-d` parameter to the `visualize` mode. +- `server` is set to either `onion` for onion service measurements or `public` for direct measurements. +- `start` is the measurement start time. +- `time_to_first_byte` is the time in seconds (with microsecond precision) to download the first byte. +- `time_to_last_byte` is the time in seconds (with microsecond precision) to download the last byte. + +## Contributing -The log indicated that port 8080 was already in use by another process. +The OnionPerf code is developed at https://gitlab.torproject.org/tpo/metrics/onionperf. -## Contribute +Contributions to OnionPerf are welcome and encouraged! -GitHub pull requests are welcome and encouraged!

1 0

[onionperf/master] Merge branch 'task-33257-2' into develop
by karsten＠torproject.org 02 Jul '20

02 Jul '20

commit 6a3ea78085532ce726a1e32413c99fc9f41a8d2b Merge: 2c19e94 4d35984 Author: Karsten Loesing <karsten.loesing(a)gmx.net> Date: Thu Jun 18 15:33:45 2020 +0200 Merge branch 'task-33257-2' into develop onionperf/visualization.py | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+)

1 0

[onionperf/master] Add CDF-DL graph.
by karsten＠torproject.org 02 Jul '20

02 Jul '20

commit 4d35984f50cb80c53c81d3b211c7aaea3551e923 Author: Karsten Loesing <karsten.loesing(a)gmx.net> Date: Thu May 28 22:46:01 2020 +0200 Add CDF-DL graph. Implements #33257. --- onionperf/visualization.py | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/onionperf/visualization.py b/onionperf/visualization.py index 339291a..2107d5a 100644 --- a/onionperf/visualization.py +++ b/onionperf/visualization.py @@ -44,6 +44,7 @@ class TGenVisualization(Visualization): self.__plot_lastbyte_box() self.__plot_lastbyte_bar() self.__plot_lastbyte_time() + self.__plot_throughput_ecdf() self.__plot_downloads_count() self.__plot_errors_count() self.__plot_errors_time() @@ -61,6 +62,16 @@ class TGenVisualization(Visualization): transfer["server"] = "onion" if ".onion:" in transfer_data["endpoint_remote"] else "public" if "elapsed_seconds" in transfer_data: s = transfer_data["elapsed_seconds"] + if "payload_progress" in s: + # Explanation of the math below for computing Mbps: From filesize_bytes + # and payload_progress fields we can compute the number of seconds that + # have elapsed between receiving bytes 524,288 and 1,048,576, which is a + # total amount of 524,288 bytes or 4,194,304 bits or 4.194304 megabits. + # We want the reciprocal of that value with unit megabits per second. + if transfer_data["filesize_bytes"] == 1048576 and "1.0" in s["payload_progress"]: + transfer["mbps"] = 4.194304 / (s["payload_progress"]["1.0"] - s["payload_progress"]["0.5"]) + if transfer_data["filesize_bytes"] == 5242880 and "0.2" in s["payload_progress"]: + transfer["mbps"] = 4.194304 / (s["payload_progress"]["0.2"] - s["payload_progress"]["0.1"]) if "first_byte" in s: transfer["time_to_first_byte"] = s["first_byte"] if "last_byte" in s: @@ -119,6 +130,13 @@ class TGenVisualization(Visualization): title="Time to download last of {0} bytes from {1} service over time".format(bytes, server), xlabel="Download start time", ylabel="Download time (s)") + def __plot_throughput_ecdf(self): + for server in self.data["server"].unique(): + self.__draw_ecdf(x="mbps", hue="label", hue_name="Data set", + data=self.data[self.data["server"] == server], + title="Throughput when downloading from {0} server".format(server), + xlabel="Throughput (Mbps)", ylabel="Cumulative Fraction") + def __plot_downloads_count(self): for bytes in np.sort(self.data["filesize_bytes"].unique()): for server in self.data["server"].unique():

1 0

[onionperf/master] Remove Sphinx documentation.
by karsten＠torproject.org 02 Jul '20

02 Jul '20

commit 145c9d1c6c281e5d2fd105d492bcc6a43d417565 Author: Karsten Loesing <karsten.loesing(a)gmx.net> Date: Tue Jun 30 17:59:44 2020 +0200 Remove Sphinx documentation. Most parts of these files, or at least the important parts, are now contained in README.md. Still part of #40001. --- onionperf/docs/Makefile | 20 ---- onionperf/docs/conf.py | 166 -------------------------- onionperf/docs/index.rst | 21 ---- onionperf/docs/onionperf.rst | 269 ------------------------------------------- 4 files changed, 476 deletions(-) diff --git a/onionperf/docs/Makefile b/onionperf/docs/Makefile deleted file mode 100644 index 91510a1..0000000 --- a/onionperf/docs/Makefile +++ /dev/null @@ -1,20 +0,0 @@ -# Minimal makefile for Sphinx documentation -# - -# You can set these variables from the command line. -SPHINXOPTS = -SPHINXBUILD = sphinx-build -SPHINXPROJ = onionperf -SOURCEDIR = . -BUILDDIR = _build - -# Put it first so that "make" without argument is like "make help". -help: - @$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O) - -.PHONY: help Makefile - -# Catch-all target: route all unknown targets to Sphinx using the new -# "make mode" option. $(O) is meant as a shortcut for $(SPHINXOPTS). -%: Makefile - @$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O) \ No newline at end of file diff --git a/onionperf/docs/conf.py b/onionperf/docs/conf.py deleted file mode 100644 index 6344c90..0000000 --- a/onionperf/docs/conf.py +++ /dev/null @@ -1,166 +0,0 @@ -# -*- coding: utf-8 -*- -# -# Configuration file for the Sphinx documentation builder. -# -# This file does only contain a selection of the most common options. For a -# full list see the documentation: -# http://www.sphinx-doc.org/en/master/config - -# -- Path setup -------------------------------------------------------------- - -# If extensions (or modules to document with autodoc) are in another directory, -# add these directories to sys.path here. If the directory is relative to the -# documentation root, use os.path.abspath to make it absolute, like shown here. -# -import os -import sys -sys.path.insert(0, os.path.abspath('..')) - - -# -- Project information ----------------------------------------------------- - -project = 'onionperf' -copyright = '2019, Ana Custura' -author = 'Ana Custura' - -# The short X.Y version -version = '' -# The full version, including alpha/beta/rc tags -release = '' - - -# -- General configuration --------------------------------------------------- - -# If your documentation needs a minimal Sphinx version, state it here. -# -# needs_sphinx = '1.0' - -# Add any Sphinx extension module names here, as strings. They can be -# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom -# ones. -extensions = [ - 'sphinx.ext.autodoc', - 'sphinx.ext.intersphinx', - 'sphinx.ext.viewcode', -] - -# Add any paths that contain templates here, relative to this directory. -templates_path = ['_templates'] - -# The suffix(es) of source filenames. -# You can specify multiple suffix as a list of string: -# -# source_suffix = ['.rst', '.md'] -source_suffix = '.rst' - -# The master toctree document. -master_doc = 'index' - -# The language for content autogenerated by Sphinx. Refer to documentation -# for a list of supported languages. -# -# This is also used if you do content translation via gettext catalogs. -# Usually you set "language" from the command line for these cases. -language = None - -# List of patterns, relative to source directory, that match files and -# directories to ignore when looking for source files. -# This pattern also affects html_static_path and html_extra_path . -exclude_patterns = ['_build', 'Thumbs.db', '.DS_Store'] - -# The name of the Pygments (syntax highlighting) style to use. -pygments_style = 'sphinx' - - -# -- Options for HTML output ------------------------------------------------- - -# The theme to use for HTML and HTML Help pages. See the documentation for -# a list of builtin themes. -# -html_theme = 'alabaster' - -# Theme options are theme-specific and customize the look and feel of a theme -# further. For a list of options available for each theme, see the -# documentation. -# -# html_theme_options = {} - -# Add any paths that contain custom static files (such as style sheets) here, -# relative to this directory. They are copied after the builtin static files, -# so a file named "default.css" will overwrite the builtin "default.css". -html_static_path = ['_static'] - -# Custom sidebar templates, must be a dictionary that maps document names -# to template names. -# -# The default sidebars (for documents that don't match any pattern) are -# defined by theme itself. Builtin themes are using these templates by -# default: ``['localtoc.html', 'relations.html', 'sourcelink.html', -# 'searchbox.html']``. -# -# html_sidebars = {} - - -# -- Options for HTMLHelp output --------------------------------------------- - -# Output file base name for HTML help builder. -htmlhelp_basename = 'onionperfdoc' - - -# -- Options for LaTeX output ------------------------------------------------ - -latex_elements = { - # The paper size ('letterpaper' or 'a4paper'). - # - # 'papersize': 'letterpaper', - - # The font size ('10pt', '11pt' or '12pt'). - # - # 'pointsize': '10pt', - - # Additional stuff for the LaTeX preamble. - # - # 'preamble': '', - - # Latex figure (float) alignment - # - # 'figure_align': 'htbp', -} - -# Grouping the document tree into LaTeX files. List of tuples -# (source start file, target name, title, -# author, documentclass [howto, manual, or own class]). -latex_documents = [ - (master_doc, 'onionperf.tex', 'onionperf Documentation', - 'Ana Custura', 'manual'), -] - - -# -- Options for manual page output ------------------------------------------ - -# One entry per manual page. List of tuples -# (source start file, name, description, authors, manual section). -man_pages = [ - (master_doc, 'onionperf', 'onionperf Documentation', - [author], 1) -] - - -# -- Options for Texinfo output ---------------------------------------------- - -# Grouping the document tree into Texinfo files. List of tuples -# (source start file, target name, title, author, -# dir menu entry, description, category) -texinfo_documents = [ - (master_doc, 'onionperf', 'onionperf Documentation', - author, 'onionperf', 'One line description of project.', - 'Miscellaneous'), -] - - -# -- Extension configuration ------------------------------------------------- - -# -- Options for intersphinx extension --------------------------------------- - -# Example configuration for intersphinx: refer to the Python standard library. -intersphinx_mapping = {'https://docs.python.org/': None} diff --git a/onionperf/docs/index.rst b/onionperf/docs/index.rst deleted file mode 100644 index e11b9a1..0000000 --- a/onionperf/docs/index.rst +++ /dev/null @@ -1,21 +0,0 @@ -.. onionperf documentation master file, created by - sphinx-quickstart on Thu Jan 24 19:39:13 2019. - You can adapt this file completely to your liking, but it should at least - contain the root `toctree` directive. - -Welcome to onionperf's documentation! -===================================== - -.. toctree:: - :maxdepth: 3 - :caption: Contents: - - onionperf - - -Indices and tables -================== - -* :ref:`genindex` -* :ref:`modindex` -* :ref:`search` diff --git a/onionperf/docs/onionperf.rst b/onionperf/docs/onionperf.rst deleted file mode 100644 index 7ef6d62..0000000 --- a/onionperf/docs/onionperf.rst +++ /dev/null @@ -1,269 +0,0 @@ - -What is OnionPerf? -================== -OnionPerf is a utility to track Tor and Onion service performance. - -OnionPerf uses multiple processes and threads to download random data through -Tor while tracking the performance of those downloads. The data is served and -fetched on localhost using two TGen (traffic generator) processes, and is -transferred through Tor using Tor client processes and an ephemeral Tor Onion -service. Tor control information and TGen performance statistics are logged to -disk, analyzed once per day to produce a json stats database and files that can -feed into Torperf, and can later be used to visualize changes in Tor client -performance over time. - -Installation -============ -OnionPerf depends on traffic generator TGen. Instructions for TGen installation can also be found at https://github.com/shadow/tgen. -To install it: - -Tgen ----- - -1. Install dependencies -:: - - apt install cmake make build-essential gcc libigraph0-dev libglib2.0-dev - -2. Clone repositories -:: - - git clone https://github.com/shadow/tgen.git - -3. Build and install -:: - - mkdir build - cd build - cmake .. -DCMAKE_INSTALL_PREFIX=/home/$USER/.local - make - ln -s build/tgen /usr/bin/tgen - - -OnionPerf ---------- -Once TGen is installed, follow these instructions to install OnionPerf: - -1. Clone repositories -:: - - apt install git - git clone https://git.torproject.org/onionperf.git - -2. Install dependencies -:: - - apt install tor libxml2-dev python-dev python-lxml python-networkx python-scipy python-matplotlib python-numpy python-netifaces python-ipaddress - -Ensure stem is the latest version for v3 Onion services to work, this can be installed from backports: -:: - - echo 'deb http://deb.debian.org/debian stretch-backports main' >> /etc/apt/sources.list - apt update - apt-get -t stretch-backports install python-stem - -3. Install OnionPerf -:: - - cd onionperf - sudo python setup.py install - -4. Run an OnionPerf test measurement -:: - - onionperf measure --oneshot --onion-only - -Watch out for any projects in the log. The output of the measurement should look similar to the following: -:: - - 2019-03-05 17:57:02 1551805022.995195 [onionperf] [INFO] Using 'tor' binary at /usr/bin/tor - 2019-03-05 17:57:02 1551805022.995359 [onionperf] [INFO] Using 'tgen' binary at /home/ana/Work/shadow/src/plugin/shadow-plugin-tgen/build/tgen - 2019-03-05 17:57:03 1551805023.005096 [onionperf] [INFO] Bootstrapping started... - 2019-03-05 17:57:03 1551805023.005265 [onionperf] [INFO] Log files for the client and server processes will be placed in /home/ana/onionperf-data - 2019-03-05 17:57:03 1551805023.005340 [onionperf] [INFO] Starting TGen server process on port 8080... - 2019-03-05 17:57:03 1551805023.012559 [onionperf] [INFO] TGen server running at 0.0.0.0:8080 - 2019-03-05 17:57:03 1551805023.012727 [onionperf] [INFO] Logging TGen server process output to /home/ana/onionperf-data/tgen-server/onionperf.tgen.log - 2019-03-05 17:57:03 1551805023.013463 [onionperf] [INFO] Starting Tor server process with ControlPort=26984, SocksPort=17674... - 2019-03-05 17:57:03 1551805023.016020 [onionperf] [INFO] Logging Tor server process output to /home/ana/onionperf-data/tor-server/onionperf.tor.log - 2019-03-05 17:57:18 1551805038.183236 [onionperf] [INFO] Logging Tor server control port monitor output to /home/ana/onionperf-data/tor-server/onionperf.torctl.log - 2019-03-05 17:57:53 1551805073.334130 [onionperf] [INFO] Creating ephemeral hidden service... - 2019-03-05 17:57:55 1551805075.847095 [onionperf] [INFO] Ephemeral hidden service is available at p3d2xcwjevqkiwtyejjbjxwadp5ces7v4k4hhrsheqwbbokuismkiyad.onion - 2019-03-05 17:57:55 1551805075.851187 [onionperf] [INFO] Starting Tor client process with ControlPort=18843, SocksPort=18397... - 2019-03-05 17:57:55 1551805075.851770 [onionperf] [INFO] Logging Tor client process output to /home/ana/onionperf-data/tor-client/onionperf.tor.log - 2019-03-05 17:58:49 1551805129.399007 [onionperf] [INFO] Logging Tor client control port monitor output to /home/ana/onionperf-data/tor-client/onionperf.torctl.log - 2019-03-05 17:58:52 1551805132.401905 [onionperf] [INFO] Starting TGen client process on port 14785... - 2019-03-05 17:58:52 1551805132.408057 [onionperf] [INFO] Logging TGen client process output to /home/ana/onionperf-data/tgen-client/onionperf.tgen.log - 2019-03-05 17:58:52 1551805132.414970 [onionperf] [INFO] Bootstrapping finished, entering heartbeat loop - 2019-03-05 17:58:53 1551805133.416607 [onionperf] [INFO] Onionperf is running in Oneshot mode. It will download a 5M file and shut down gracefully... - 2019-03-05 17:59:24 1551805164.697978 [onionperf] [INFO] Onionperf has downloaded a 5M file in oneshot mode, and will now shut down. - 2019-03-05 17:59:24 1551805164.698091 [onionperf] [INFO] Cleaning up child processes now... - 2019-03-05 17:59:24 1551805164.707111 [onionperf] [INFO] Joining tgen_server_watchdog thread... - 2019-03-05 17:59:25 1551805165.094310 [onionperf] [INFO] Joining tor_server_watchdog thread... - 2019-03-05 17:59:25 1551805165.094690 [onionperf] [INFO] Joining torctl_server_helper thread... - 2019-03-05 17:59:25 1551805165.113840 [onionperf] [INFO] command '/home/ana/Work/shadow/src/plugin/shadow-plugin-tgen/build/tgen /home/ana/onionperf-data/tgen-client/tgen.graphml.xml' finished as expected - 2019-03-05 17:59:25 1551805165.493231 [onionperf] [INFO] Joining tor_client_watchdog thread... - 2019-03-05 17:59:25 1551805165.493633 [onionperf] [INFO] Joining torctl_client_helper thread... - 2019-03-05 17:59:25 1551805165.538060 [onionperf] [INFO] Joining tgen_client_watchdog thread... - 2019-03-05 17:59:25 1551805165.538464 [onionperf] [INFO] Joining logrotate thread... - 2019-03-05 17:59:26 1551805166.539878 [onionperf] [INFO] Child processes terminated - 2019-03-05 17:59:26 1551805166.540305 [onionperf] [INFO] Child process cleanup complete! - 2019-03-05 17:59:26 1551805166.540603 [onionperf] [INFO] Exiting - -Deployment -========== - -Deployment options ------------------- -There are -various options available for deployment of measurements. These options can be -passed to OnionPerf via command line arguments, as follows: - -:: - - --onion-only - -Only download files through an Onion service (default: disabled) :: - - --inet-only - -Only download files through the Internet (default: disabled) :: - - --single-onion - -Use a single Onion service, which uses a direct circuit between the Onion -service and the introduction and rendezvous point. :: - - --torclient-conf-file FILE - -Download files using specified configuration file for the Tor client (default: disabled) :: - - --torserver-conf-file FILE - -In addition to specifying configuration files, you can pass newline-separated -Tor configuration options to the Tor process by adding them to the -:code:`BASETORRC` environment variable. These options are prepended to all other -configuration options. Here is an example which prepends the options -:code:`Option1 Foo` and :code:`Option2 Bar` to Tor's configuration file: :: - - BASETORRC=$'Option1 Foo\nOption2 Bar' onionperf ... - -Download files using specified configuration file for the Tor server (default: disabled) :: - - --additional-client-conf STRING - -Download files using specified configuration lines (default: disabled) - -By default, OnionPerf downloads files using both the Internet and Onion services, using a v3 Onion address. -It uses publicly available relays, but by specifying additional configuration files it can be configured to run -on test Tor networks, or using bridges with or without pluggable transports. -:: - - --oneshot - -Only download a 5M file and then shut down gracefully (default: disabled) - -By default, OnionPerf runs continuously and appends measurement information to -log files as they happen. At midnight, the log files are rotated and the measurement continues. -A oneshot measurement will run only until one successful download has completed. - -:: - - --nickname STRING - -The 'SOURCE' STRING to use in stats files produced by OnionPerf (default: hostname of the current machine) -:: - - --prefix PATH - -A directory PATH prefix where OnionPerf will run (default: current directory) -:: - - --tor PATH - -A file PATH to a Tor binary (default: looks in $PATH) -:: - - --tgen PATH - -A file PATH to a TGen binary (default: looks in $PATH) - -Example vanilla Tor deployment ------------------------------- - -The following command will download files continuously using a Tor client through Onion service version 3 and via the Internet until it is stopped: -:: - - onionperf --measure - - -Example vanilla bridge deployment ---------------------------------- -The following command will download files continuously using a Tor client through Onion service version 3 and via the Internet until it is stopped. -The Tor client will always pick one of the bridges provided in this configuration to be the first hop in the circuits it builds: - -:: - - onionperf --measure --additional-client-conf="UseBridges 1 - Bridge 72.14.177.231:9001 AC0AD4107545D4AF2A595BC586255DEA70AF119D - Bridge 195.91.239.8:9001 BA83F62551545655BBEBBFF353A45438D73FD45A - Bridge 148.63.111.136:35577 768C8F8313FF9FF8BBC915898343BC8B238F3770" - -Note: a new line must be added at the end of each configuration directive. - -A second way of passing this configuration to OnionPerf would be to create a file called tor_conf in a directory of your choice, containing the lines: -:: - - UseBridges 1 - Bridge 148.63.111.136:35577 768C8F8313FF9FF8BBC915898343BC8B238F3770 - Bridge 195.91.239.8:9001 BA83F62551545655BBEBBFF353A45438D73FD45A - Bridge 148.63.111.136:35577 768C8F8313FF9FF8BBC915898343BC8B238F3770 - -This file is then passed to the client configurator in OnionPerf: - -:: - - onionperf --measure --torclient-config-file=/path/to/tor_conf - -If we want to use vanilla Tor for the client, but download the files through an Onion service accessible via a bridge, the same configuration file containing the bridge lines can be passed to the server: - -:: - - onionperf --measure --torserver-config-file=/path/to/tor_conf - - -Note that bridge lines for configuration can be downloaded from https://bridges.torproject.org. - -Example bridge with Pluggable Transport deployment --------------------------------------------------- -Similarly to the above, the Tor client can use Pluggable Transports (PT) with bridges. Here we present examples for meek and obfs4proxy. - -You must have the meek and/or obfs4proxy binaries installed. The binaries can -be obtained by downloading the latest version of Tor browser bundle, or they -can be installed from source. In the example file that follows, directive "ClientTransportPlugin" -needs to point to the path of the binary corresponding to the wanted PT. Finally, both meek and -obfs4 enabled bridges can be obtained from the bridge database. - -Example file torrc1: -:: - - UseBridges 1 - # Example meek bridge line - meek bridge lines can be downloaded from https://bridges.torproject.org - Bridge meek 0.0.2.0:1 url=https://at-b2.erg.abdn.ac.uk - # meek configuration - ClientTransportPlugin meek exec /usr/bin/meek - -Example file torrc2: -:: - - # Example obfs4 bridge - meek bridge lines can be downloaded from bridges.torproject.org - Bridge obfs4 137.50.19.19:5001 AE77C35CAC66C2F207319939029D6D22945BDA84 cert=kwpT6sHRa80CnoSCGzelo2wl4RU7cC+mjBCihj2gAJAnvNyTWD3Pk9Ae05+fGpiGzHleOw iat-mode=0 - # obfs4 configuration - ClientTransportPlugin obfs2,obfs3,obfs4,scramblesuit exec /usr/bin/obfs4proxy - -Then, the configuration files containing the required bridge and PT lines can be passed to the either the Tor server or client: - -:: - - onionperf --measure --torserver-config-file=/path/to/torrc1 --torclient-config-file=/path/to/torrc2 - -In this example, the Onion services uses the obfs4 bridge configured in file torrc2 to connect to the Tor network, while the client uses the meek bridge configured in file torrc1 to connect to the Tor network.

1 0

[onionperf/master] Always include error_code column in CSV output.
by karsten＠torproject.org 02 Jul '20

02 Jul '20

commit fc6e9702d3012edca463b96fbbe218bbff70d6c7 Author: Karsten Loesing <karsten.loesing(a)gmx.net> Date: Wed Jul 1 15:20:27 2020 +0200 Always include error_code column in CSV output. We should include an error_code column in the CSV output, regardless of whether data contains measurements with an error code or not. Everything else is just too confusing to whoever processes these files. Fixes #40004. --- onionperf/visualization.py | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/onionperf/visualization.py b/onionperf/visualization.py index 2107d5a..0a2a9d9 100644 --- a/onionperf/visualization.py +++ b/onionperf/visualization.py @@ -58,7 +58,8 @@ class TGenVisualization(Visualization): tgen_transfers = analysis.get_tgen_transfers(client) for transfer_id, transfer_data in tgen_transfers.items(): transfer = {"transfer_id": transfer_id, "label": label, - "filesize_bytes": transfer_data["filesize_bytes"]} + "filesize_bytes": transfer_data["filesize_bytes"], + "error_code": None} transfer["server"] = "onion" if ".onion:" in transfer_data["endpoint_remote"] else "public" if "elapsed_seconds" in transfer_data: s = transfer_data["elapsed_seconds"] @@ -147,7 +148,7 @@ class TGenVisualization(Visualization): def __plot_errors_count(self): for server in self.data["server"].unique(): - if "error_code" in self.data.columns and self.data[self.data["server"] == server]["error_code"].count() > 0: + if self.data[self.data["server"] == server]["error_code"].count() > 0: self.__draw_countplot(x="error_code", hue="label", hue_name="Data set", data=self.data[self.data["server"] == server], xlabel="Error code", ylabel="Downloads failed (#)", @@ -155,7 +156,7 @@ class TGenVisualization(Visualization): def __plot_errors_time(self): for server in self.data["server"].unique(): - if "error_code" in self.data.columns and self.data[self.data["server"] == server]["error_code"].count() > 0: + if self.data[self.data["server"] == server]["error_code"].count() > 0: self.__draw_stripplot(x="start", y="error_code", hue="label", hue_name="Data set", data=self.data[self.data["server"] == server], xlabel="Download start time", ylabel="Error code",

1 0

[onionperf/master] Merge branch 'task-40002' into develop
by karsten＠torproject.org 02 Jul '20

02 Jul '20

commit 0a9c6aca86f44660e4895c643484c0c16fc9c705 Merge: 3e1d149 30949a4 Author: Karsten Loesing <karsten.loesing(a)gmx.net> Date: Thu Jul 2 10:22:24 2020 +0200 Merge branch 'task-40002' into develop onionperf/measurement.py | 3 +++ 1 file changed, 3 insertions(+)

1 0

[onionperf/master] Write generated torrc files to disk.
by karsten＠torproject.org 02 Jul '20

02 Jul '20

commit 30949a4218cf06c8e73927ec0db06dfa6d662294 Author: Karsten Loesing <karsten.loesing(a)gmx.net> Date: Wed Jul 1 16:50:57 2020 +0200 Write generated torrc files to disk. Implements #40002. --- onionperf/measurement.py | 3 +++ 1 file changed, 3 insertions(+) diff --git a/onionperf/measurement.py b/onionperf/measurement.py index 4a58bc4..1540bac 100644 --- a/onionperf/measurement.py +++ b/onionperf/measurement.py @@ -439,6 +439,9 @@ WarnUnsafeSocks 0\nSafeLogging 0\nMaxCircuitDirtiness 60 seconds\nDataDirectory if not os.path.exists(tor_datadir): os.makedirs(tor_datadir) tor_config = self.create_tor_config(control_port,socks_port,tor_datadir,name) + tor_confpath = "{0}/torrc".format(tor_datadir) + with open(tor_confpath, 'wt') as f: + f.write(tor_config) tor_logpath = "{0}/onionperf.tor.log".format(tor_datadir) tor_writable = util.FileWritable(tor_logpath)

1 0

[onionperf/master] Rewrites/clarifies some text
by karsten＠torproject.org 02 Jul '20

02 Jul '20

commit 3c129c371a10a69a1789c1af4ac8495a0a783203 Author: Ana Custura <ana(a)netstat.org.uk> Date: Thu Jul 2 10:50:29 2020 +0100 Rewrites/clarifies some text --- README.md | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/README.md b/README.md index 7ffb900..1f58f0d 100644 --- a/README.md +++ b/README.md @@ -36,15 +36,15 @@ OnionPerf does not interfere with how Tor selects paths and builds circuits, oth ## Installation -OnionPerf has a couple dependencies in order to perform measurements or analyze and visualize measurement results. These dependencies include Tor, TGen (traffic generator), and a couple Python packages. +OnionPerf has several dependencies in order to perform measurements or analyze and visualize measurement results. These dependencies include Tor, TGen (traffic generator), and a few Python packages. The following description was written with a Debian system in mind but should be transferable to other Linux distributions and possibly even other operating systems. ### Tor -OnionPerf relies on the `tor` binary to start a Tor client on the client side and another one on the server side to host onion services. +OnionPerf relies on the `tor` binary to start a Tor client process on the client side and a server side process to host onion services. -The easiest way is to install the `tor` package, which puts the `tor` binary into the `PATH` where OnionPerf will find it. Optionally, systemd can be instructed to make sure that `tor` is never started as a service: +The easiest way to satisfy this dependency is to install the `tor` package, which puts the `tor` binary into the `PATH` where OnionPerf will find it. Optionally, systemd can be instructed to make sure that `tor` is never started as a service: ```shell sudo apt install tor @@ -88,7 +88,7 @@ The TGen binary will be contained in `~/tgen/build/tgen`, which is also the path OnionPerf is written in Python 3. The following instructions assume that a Python virtual environment is being used, even though installation is also possible without that. -The virtual environment is being created, activated, and tested using: +The virtual environment is created, activated, and tested using: ```shell sudo apt install python3-venv @@ -100,14 +100,14 @@ which python3 The last command should output something like `~/venv/bin/python3` as the path to the `python3` binary used in the virtual environment. -In the next step, the OnionPerf repository is being cloned and requirements are being installed: +The next step is to clone the OnionPerf repository and install its requirements: ```shell git clone https://git.torproject.org/onionperf.git pip3 install --no-cache -r onionperf/requirements.txt ``` -The next step after that is to install OnionPerf and print out the usage information to see if the installation was successful: +The final step is to install OnionPerf and print out the usage information to see if the installation was successful: ```shell cd onionperf/ @@ -122,7 +122,7 @@ The virtual environment is deactivated with the following command: deactivate ``` -However, in order to perform measurements or doing analyses, the virtual environment needs to be activated first. This will ensure all the paths are found. +However, in order to perform measurements or analyses, the virtual environment needs to be activated first. This will ensure all the paths are found. ## Measurement @@ -133,7 +133,7 @@ Ideally, OnionPerf is run detached from the terminal session using tmux, systemd ### Starting and stopping measurements -The most trivial configuration is to measure onion services only. In that case, OnionPerf is taking care of all configurations, and there are no firewall rules or port forwards to take care of. +The most trivial configuration is to measure onion services only. In that case, OnionPerf runs without needing any additional configuration. For direct measurements via exit nodes, firewall rules or port forwarding may be required to allow inbound connections to the TGen server. Starting these measurements is as simple as:

1 0

[onionperf/master] Merge branch 'task-40004' into develop
by karsten＠torproject.org 02 Jul '20

02 Jul '20

commit 83e4b52575576806f1879f3912f4a913c3e06747 Merge: 0a9c6ac fc6e970 Author: Karsten Loesing <karsten.loesing(a)gmx.net> Date: Thu Jul 2 10:43:05 2020 +0200 Merge branch 'task-40004' into develop onionperf/visualization.py | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)

1 0

[onionperf/master] Minor spelling, adds TOC
by karsten＠torproject.org 02 Jul '20

02 Jul '20

commit 508d3bccedc56af6559bcb5e367172dffec73a05 Author: Ana Custura <ana(a)netstat.org.uk> Date: Thu Jul 2 10:02:47 2020 +0100 Minor spelling, adds TOC --- README.md | 26 ++++++++++++++++++++++---- 1 file changed, 22 insertions(+), 4 deletions(-) diff --git a/README.md b/README.md index 9f0faf4..b66d6bc 100644 --- a/README.md +++ b/README.md @@ -1,12 +1,30 @@ # OnionPerf - -[TOC] + * [Overview](#overview) + + [What does OnionPerf do?](#what-does-onionperf-do-) + + [What does OnionPerf *not* do?](#what-does-onionperf--not--do-) + * [Installation](#installation) + + [Tor](#tor) + + [TGen](#tgen) + + [OnionPerf](#onionperf-1) + * [Measurement](#measurement) + + [Starting and stopping measurements](#starting-and-stopping-measurements) + + [Output directories and files](#output-directories-and-files) + + [Changing Tor configurations](#changing-tor-configurations) + + [Changing the TGen traffic model](#changing-the-tgen-traffic-model) + + [Sharing measurement results](#sharing-measurement-results) + + [Troubleshooting](#troubleshooting) + * [Analysis](#analysis) + + [Analyzing measurement results](#analyzing-measurement-results) + + [Visualizing measurement results](#visualizing-measurement-results) + + [Interpreting the PDF output format](#interpreting-the-pdf-output-format) + + [Interpreting the CSV output format](#interpreting-the-csv-output-format) + * [Contributing](#contributing) ## Overview ### What does OnionPerf do? -OnionPerf measures performance of bulk file downloads over Tor. Together with its predecessor, Torperf, Onionperf has been used to measure long-term performance trends in the Tor network since 2009. It is also being used to perform short-term performance experiments to compare different Tor configurations or implementations. +OnionPerf measures performance of bulk file downloads over Tor. Together with its predecessor, Torperf, OnionPerf has been used to measure long-term performance trends in the Tor network since 2009. It is also being used to perform short-term performance experiments to compare different Tor configurations or implementations. OnionPerf uses multiple processes and threads to download random data through Tor while tracking the performance of those downloads. The data is served and fetched on localhost using two TGen (traffic generator) processes, and is transferred through Tor using Tor client processes and an ephemeral Tor onion service. Tor control information and TGen performance statistics are logged to disk and analyzed once per day to produce a JSON analysis file that can later be used to visualize changes in Tor client performance over time. @@ -180,7 +198,7 @@ For example, the hard-coded traffic model generated by OnionPerf and executed by ### Sharing measurement results -Measurement results can be further analyzed and visualized on the measuring host. But in many cases it's more conventient to do analysis and visualization on another host, also to compare measurements from different hosts to each other. +Measurement results can be further analyzed and visualized on the measuring host. But in many cases it's more convenient to do analysis and visualization on another host, also to compare measurements from different hosts to each other. There are at least two common ways of sharing measurement results:

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

tor-commits July 2020