Archiveopteryx

[Nod]

Are anybody using Archiveopteryx?

Archiveopteryx is an Internet mail server, optimised to support long-term archival storage. It seeks to make it practical not only to manage large archives, but to use the information therein on a daily basis instead of relegating it to offline storage.

Archiveopteryx runs on Linux, FreeBSD, NetBSD, OpenBSD and Mac OS X. The source code is written in core C++, using very few libraries, and compiles without any warnings.

Archiveopteryx is designed to impose no limits on the size or usage of the archive to the extent of the server hardware's capabilities. It supports a clustered configuration to deal with higher load, and needs no ad-hoc workarounds for common limitations on, for example, the number or size of folders.
Searching and categorisation

Archiveopteryx uses an RDBMS to store mail in a strictly normalized format to make retrieval easy, and to avoid duplication. This makes it possible to search through and categorise even very large quantities of mail in ways that were not anticipated when the archive was first created.

We support all of IMAP's bandwidth-saving features, including partial fetches and compression, so good clients can access a large mailbox using very little bandwidth.

 

[dequadin]

No but looks interesting.

I started writing something similar last year (for a project that got shelved) and the 4000 BiLLiOn different MIME types and encoding schemes in the RFC's make it quite a daunting task.

If this can handle all that grunt work that's awesome...

 

[Nod]

I think run a test to see how it works. Might be a nice experiment.

Just compiled it. No issues as advertised.