MirrorBrain is a framework to run a content delivery network using mirror servers. It solves a challenge that many popular open source projects face: a flood of download requests, often magnitudes more than a single site could practically handle. A central (and probably the most obvious) part is a "download redirector" that automatically redirects requests from Web browsers or download programs to a mirror server near them. Choosing a suitable mirror for a user's request is the key, and MirrorBrain uses geolocation and global routing data to make a sensible choice and achieve load-balancing for the mirrors at the same time. The algorithm is both sophisticated and easy to control and tune. In addition, MirrorBrain monitors mirrors, scans them for files, generates mirror lists, and more.
Why a milter archiver? If you are required to also log source and destination IPs and BCCs (additional rcpt to's not in a CC field communicated at the SMTP protocol level), then the only option is do this in the MTA, or in a milter application that has hooks in the MTA process (which is exactly why milter was invented). It uses a simple regex file that can be used with several *source target-mailbox rules to define what gets mirrored where. The archiver tries to be as quiet as possible. The possible methods to archive are SMTP, sendmail inject / Postfix inject, IMAP delivery, and milter-add recipient. It was build as a replacement for a Postfix BCC setup: (sender_bcc_maps = regexp:/etc/postfix/archive; recipient_bcc_maps = regexp:/etc/postfix/archive).