Welcome to Allmydata "Tahoe"

The ​Allmydata "Tahoe" project is a distributed, fault-tolerant storage mesh, available under a free-software license.

You can store files of any length on the mesh and retrieve them with a cryptographic identifier that securely identifies that file. You can also organize files with filenames and directories, in the familiar style.

The basic idea is that the data in this storage mesh is spread over all participating nodes, using an algorithm that can recover the data even if a majority of the nodes are no longer available.

The code is currently functional, but it is targeted at hackers who are willing to build from source and use a minimal, text-oriented web user interface.

Please see ​the README file and ​the release notes for an introduction.

Tahoe resources

• The code is available via darcs by running the following command:

  darcs get http://allmydata.org/source/tahoe/trunk
  

• Other related libraries are available in ​http://allmydata.org/source/
• ​Buildbot: current build status
• Archived code coverage data (scroll down for the most recent build)
• Graph of code coverage
• The architectural overview documents some of the design.
• to install debian packages for etch, dapper, edgy, or feisty, add the following lines to your /etc/apt/sources.list, replacing the word DIST with etch, dapper, edgy, or feisty as appropriate:

  deb http://allmydata.org/debian/  DIST  main tahoe
  deb-src http://allmydata.org/debian/  DIST  main tahoe
  

  Then update and install the allmydata-tahoe package. These packages are generated each time the code is changed, and represent the most up-to-date (read "unstable") version available. The usual warnings about no guarantees apply: it might cause your computer to catch fire, might steal your dog, etc.
  • we need to set up an etch buildslave, as we don't currently have one running
• TahoeIssues? contains some discussion, open issues, and more docs
• The tahoe-dev mailing list is the best place to discuss the project.
  • The zfec-dev mailing list is specific to the zfec erasure-coding library.

License

Tahoe is offered under the GNU General Public License (v2). It also comes with the added permission that, in the case that you are obligated to release a derived work under this licence (as per section 2.b of the GPLv2), you may delay the fulfillment of this obligation for up to 12 months.

Supporting Libraries

This project depends upon several others. For full details please consult the README, but here is a quick list of things that you will need to install before running a tahoe client or introducer:

• Python (​http://python.org), version 2.4 or 2.5
• Twisted (​http://twistedmatrix.com), we run with 2.4.x or 2.5.x
• Foolscap (​http://twistedmatrix.com/trac/wiki/FoolsCap), we require 0.1.2
• PyOpenSSL (​http://pyopenssl.sourceforge.net), 0.6 or later
• Nevow (​http://divmod.org/trac/wiki/DivmodNevow) for the web frontend, we use 0.9.x
• zfec (bundled with the tahoe tree)
• a patched PyCrypto? (also bundled with the tahoe tree, installs into allmydata.Crypto)

Test Network

An experimental public testnet is running on Allmydata servers. It is subject to being updated at random times, so compatibility is like to break without notice.

After installing the code according to the README, set up a client node with the following files in its base directory:

file name	file contents
introducer.furl	pb://xextf3eap44o3wi27mf7ehiur6wvhzr6@207.7.153.180:56677,127.0.0.1:56677/introducer
vdrive.furl	pb://xextf3eap44o3wi27mf7ehiur6wvhzr6@207.7.153.180:56677,127.0.0.1:56677/vdrive

The testnet is currently running foolscap-0.1.2, available in ​http://allmydata.org/source/ . All nodes in the network must be running the same version of Foolscap to be able to establish connections.

Trac Starting Points

• OriginalWikiStart -- the original Trac welcome page
• TracGuide -- Built-in Documentation
• ​The Trac project -- Trac Open Source Project
• ​Trac FAQ -- Frequently Asked Questions
• TracSupport -- Trac Support

For a complete list of local wiki pages, see TitleIndex.