brozzler - distributed browser-based web crawler
Python JavaScript HTML Other
Latest commit a1f1681 Feb 24, 2017 @nlevitt nlevitt fix issue where use of YoutubeDLSpy caused youtube-dl connections to …
…remote servers to be kept open


logo brozzler

"browser" | "crawler" = "brozzler"

Brozzler is a distributed web crawler (爬虫) that uses a real browser (chrome or chromium) to fetch pages and embedded urls and to extract links. It also uses youtube-dl to enhance media capture capabilities.

Brozzler is designed to work in conjunction with warcprox for web archiving.


  • Python 3.4 or later
  • RethinkDB deployment
  • Chromium or Google Chrome browser

Worth noting is that the browser requires a graphical environment to run. You already have this on your laptop, but on a server it will probably require deploying some additional infrastructure (typically X11). The vagrant configuration in the brozzler repository (still a work in progress) has an example setup.

Getting Started

The easiest way to get started with brozzler for web archiving is with brozzler-easy. Brozzler-easy runs brozzler-worker, warcprox, pywb, and brozzler-dashboard, configured to work with each other, in a single process.

Mac instructions:

# install and start rethinkdb
brew install rethinkdb
# no brew? try rethinkdb's installer:
rethinkdb &>>rethinkdb.log &

# install brozzler with special dependencies pywb and warcprox
pip install brozzler[easy]  # in a virtualenv if desired

# queue a site to crawl

# or a job
brozzler-new-job job1.yml

# start brozzler-easy

At this point brozzler-easy will start brozzling your site. Results will be immediately available for playback in pywb at http://localhost:8880/brozzler/.

Brozzler-easy demonstrates the full brozzler archival crawling workflow, but does not take advantage of brozzler's distributed nature.

Installation and Usage

To install brozzler only:

pip install brozzler  # in a virtualenv if desired

Launch one or more workers:


Submit jobs:

brozzler-new-job myjob.yaml

Submit sites not tied to a job:

brozzler-new-site --proxy=localhost:8000 --enable-warcprox-features \

Job Configuration

Jobs are defined using yaml files. Options may be specified either at the top-level or on individual seeds. At least one seed url must be specified, everything else is optional. For details, see job-conf.rst.

id: myjob
time_limit: 60 # seconds
proxy: # point at warcprox for archiving
ignore_robots: false
enable_warcprox_features: false
warcprox_meta: null
metadata: {}
  - url:
  - url:
    time_limit: 30
  - url:
    time_limit: 10
    ignore_robots: true
      surt: http://(org,example,

Brozzler Dashboard

Brozzler comes with a rudimentary web application for viewing crawl job status. To install the brozzler with dependencies required to run this app, run

pip install brozzler[dashboard]

To start the app, run


See brozzler-dashboard --help for configuration options.

Brozzler Wayback

Brozzler comes with a customized version of pywb which supports using the rethinkdb "captures" table (populated by warcprox) as its index.

To use, first install dependencies.

pip install brozzler[easy]

Write a configuration file pywb.yml.

# 'archive_paths' should point to the output directory of warcprox
archive_paths: warcs/  # pywb will fail without a trailing slash
    index_paths: !!python/object:brozzler.pywb.RethinkCDXSource
      db: brozzler
      table: captures
      - localhost
enable_auto_colls: false
enable_cdx_api: true
framed_replay: true
port: 8880

Run pywb like so:

$ PYWB_CONFIG_FILE=pywb.yml brozzler-wayback

Then browse http://localhost:8880/brozzler/.

Headless Chromium

Headless Chromium may optionally be used instead of Chromium or Chrome to run Brozzler without a visisble browser window or X11 server. At the time of writing headless_shell is a separate Linux-only executable and must be compiled from source. Beware that compiling Chromium requires 10 GB of disk space, several GB of RAM and patience.

Start by installing the dependencies listed in Chromium's Linux-specific build instructions.

Next install the build tools and fetch the source code:

mkdir -p ~/chromium
cd ~/chromium
git clone
export PATH=$PWD/depot_tools:$PATH
fetch --no-history chromium --nosvn=True

Configure a headless release build (the debug builds are much larger):

cd src
mkdir -p out/release
echo 'import("//build/args/")' > out/release/
echo 'is_debug = false' >> out/release/
gn gen out/release

Run the compile:

ninja -C out/release headless_shell

This will produce an out/release/headless_shell executable. Unfortunately this cannot be used with Brozzler as-is as the --window-size command-line option expects a different syntax in Headless Chromium. As a workaround create a wrapper shell script which replaces the misbehaving option:

exec ~/chromium/src/out/release/headless_shell "${@//--window-size=1100,900/--window-size=1100x900}"

Run brozzler passing the path to the wrapper script as the --chrome-exe option:

chmod +x ~/bin/
brozzler-worker --chrome-exe ~/bin/

To render Flash content, download and extract the Linux (.tar.gz) PPAPI plugin. Configure Headless Chromium to load the plugin by adding this option to your wrapper script:



Copyright 2015-2017 Internet Archive

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this software except in compliance with the License. You may obtain a copy of the License at

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.