NewsGrabber-Warrior

More information about the archiving project can be found on the ArchiveTeam wiki: NewsGrabber

Setup instructions

There are now several ways to run this; the preffered method is via the included Dockerfile.

Be sure to replace YOURNICKHERE with the nickname that you want to be shown as, on the tracker. You don't need to register it, just pick a nickname you like.

In most of the below cases (with the exception of docker by default), there will be a web interface running at http://localhost:8001/. If you don't know or care what this is, you can just ignore it—otherwise, it gives you a fancy view of what's going on.

If anything goes wrong while running the commands below, please scroll down to the bottom of this page. There's troubleshooting information there.

Running with docker

Assuming this is a stand alone box, not part of a swarm etc, basic instructions for configuring your docker instance can be found at docker documentation or for Ubuntu / Debian.

Make a directory, cd into the directry and copy the included dockerfile into it; the rest of the files are not required. Edit the final line to include the concurrency (CPU bound due to the deduplication, recommend 1.5 times CPU / vCPU) and replace UnknownDocker with your username.

Build the container with the following arguments;

docker build -t <<dockername>> <<foldername>>/

for example

docker build -t newsgrabber-warrior newsgrabber-warrior/

Then simply run the container with either;

docker run -d -it newsgrabber-warrior

or if you want to give it a known name and make it easier to run commands;

docker run -d -it --name newsgrabber-warrior newsgrabber-warrior

or if you really want that web page to be available;

docker run -d -it -p 8001:8001 --name newsgrabber-warrior newsgrabber-warrior

Stopping the container (clean);

docker run -d -it --name <<containername>> touch STOP

Stopping the container (hard);

docker stop <<containername>>

Connecting to the container console;

docker attach <<containername>>

Running with a warrior

Follow the instructions on the ArchiveTeam wiki for installing the Warrior, and select the "NewsGrabber" project in the Warrior interface.

Running without a warrior

To run this outside the warrior, clone this repository, cd into its directory and run:

pip install --upgrade seesaw

then start downloading with:

run-pipeline pipeline.py --concurrent 2 YOURNICKHERE

For more options, run:

run-pipeline --help

If you don't have root access and/or your version of pip is very old, you can replace "pip install --upgrade seesaw" with:

wget https://raw.github.com/pypa/pip/master/contrib/get-pip.py ; python get-pip.py --user ; ~/.local/bin/pip install --user seesaw

so that pip and seesaw are installed in your home, then run

~/.local/bin/run-pipeline pipeline.py --concurrent 2 YOURNICKHERE

Running multiple instances on different IPs

This feature requires seesaw version 0.0.16 or greater. Use pip install --upgrade seesaw to upgrade.

Use the --context-value argument to pass in bind_address=123.4.5.6 (replace the IP address with your own).

Example of running 2 threads, no web interface, and Wget binding of IP address:

run-pipeline pipeline.py --concurrent 2 YOURNICKHERE --disable-web-server --context-value bind_address=123.4.5.6

Distribution-specific setup

For Debian/Ubuntu:

adduser --system --group --shell /bin/bash archiveteam
apt-get update && apt-get install -y git-core libgnutls-dev screen python-dev python-pip bzip2 zlib1g-dev unzip
pip install --upgrade seesaw requests warcio dnspython
su -c "cd /home/archiveteam; git clone https://github.com/ArchiveTeam/NewsGrabber-Warrior.git" archiveteam
su -c "cd /home/archiveteam/NewsGrabber-Warrior/; wget https://launchpad.net/wpull/trunk/v1.2.3/+download/wpull-1.2.3-linux-x86_64-3.4.3-20160302011013.zip; unzip wpull-1.2.3-linux-x86_64-3.4.3-20160302011013.zip; chmod +x ./wpull" archiveteam
screen su -c "cd /home/archiveteam/NewsGrabber-Warrior/; run-pipeline pipeline.py --concurrent 2 --address '127.0.0.1' YOURNICKHERE" archiveteam
[... ctrl+A D to detach ...]

For CentOS:

Ensure that you have the CentOS equivalent of bzip2 installed as well. You might need the EPEL repository to be enabled.

yum -y install gnutls-devel python-pip zlib-devel unzip
pip install --upgrade seesaw requests warcio dnspython
[... pretty much the same as above ...]

For openSUSE:

zypper install screen python-pip libgnutls-devel bzip2 python-devel gcc make unzip
pip install --upgrade seesaw requests warcio dnspython
[... pretty much the same as above ...]

For OS X:

You need Homebrew. Ensure that you have the OS X equivalent of bzip2 installed as well.

brew install python gnutls unzip
pip install --upgrade seesaw requests warcio dnspython
[... pretty much the same as above ...]

There is a known issue with some packaged versions of rsync. If you get errors during the upload stage, NewsGrabber-Warrior will not work with your rsync version.

This supposedly fixes it:

alias rsync=/usr/local/bin/rsync

For Arch Linux:

Ensure that you have the Arch equivalent of bzip2 installed as well.

1. Make sure you have python2-pip installed.
2. Run pip2 install seesaw.
3. Modify the run-pipeline script in seesaw to point at #!/usr/bin/python2 instead of #!/usr/bin/python.
4. useradd --system --group users --shell /bin/bash --create-home archiveteam
5. su -c "cd /home/archiveteam; git clone https://github.com/ArchiveTeam/NewsGrabber-Warrior.git" archiveteam
6. su -c "cd /home/archiveteam/NewsGrabber-Warrior/; wget https://launchpad.net/wpull/trunk/v1.2.3/+download/wpull-1.2.3-linux-x86_64-3.4.3-20160302011013.zip; unzip wpull-1.2.3-linux-x86_64-3.4.3-20160302011013.zip; chmod +x ./wpull" archiveteam
7. screen su -c "cd /home/archiveteam/NewsGrabber-Warrior/; run-pipeline pipeline.py --concurrent 2 --address '127.0.0.1' YOURNICKHERE" archiveteam

For FreeBSD:

Nothing specific here. If not so, please do let us know on IRC (irc.efnet.org #archiveteam).

Troubleshooting

Broken? These are some of the possible solutions:

Wpull not successfully running

If you have trouble getting Wpull running, please see http://wpull.readthedocs.org/en/master/install.html.

Problem with gnutls or openssl during building

Please ensure that gnutls-dev(el) and openssl-dev(el) are installed.

ImportError: No module named seesaw

If you're sure that you followed the steps to install seesaw, permissions on your module directory may be set incorrectly. Try the following:

chmod o+rX -R /usr/local/lib/python2.7/dist-packages

run-pipeline: command not found

Install seesaw using pip2 instead of pip.

pip2 install seesaw

Issues in the code

If you notice a bug and want to file a bug report, please use the GitHub issues tracker.

Are you a developer? Help write code for us! Look at our developer documentation for details.

Other problems

Have an issue not listed here? Join us on IRC and ask! We can be found at irc.efnet.org #newsgrabber.