Next Previous Contents

1. Introduction

Normal POP mail sessions, by their very nature, are insecure. The password goes across the network in cleartext for everyone to see. Now, this may be perfectly acceptable in a trusted or firewalled environment. But on a public network, such as a university or your run-of-the-mill ISP, anyone armed with a simple network sniffer can grab your password right off the wire. This is compounded by the fact that many people set their computers to check for mail at regular intervals, so the password is sent out quite frequently, which makes it easy to sniff.

With this password, an attacker can now access your email account, which may have sensitive or private information. It is also quite common that this password is the same as the user's shell account, so there is the possibility for more damage.

By doing all POP traffic using an encrypted channel, nothing goes in cleartext over the network. We can use ssh's diverse methods of authentication, instead of a simple plaintext password. That is the real point of using this method: not because we get encrypted content (which is futile at this point, since it's probably gone unencrypted over several networks already before reaching your mailbox; securing those communications is the job of GNU Privacy Guard or PGP, not ssh), but the secure authentication.

There are other methods of achieving secure authentication already, such as APOP, KPOP, and IMAP. However, using ssh has the advantage that it works with normal POP configurations, without requiring special client (not all mail clients support advanced protocols) or server support (except for sshd running on the server). You mail provider may be unable or unwilling to use a more secure protocol. Besides, by using ssh you can compress the traffic too, which is a nice little extra for people with slow connections.

Next Previous Contents