It started with a short spam apparently advertising a fairly well known black hat forum for hackers and illicit trades. It's not the sort of place that would choose to advertise itself though (it is strictly by invitation only), so quite possibly this is a Joe Job by one set of black hatters against another.
Now I guess that many recipients will have done the same thing, and typed the name of the site into Google to find out about it.. under the assumption that they'll find something that doesn't involve visiting the spamvertised site itself. But if you're using Firefox (and this possibly applies to IE8 and IE9 too, then the following message pops up:
Secure Connection FailedRight at this point I kicked myself because I thought I had accidentally clicked through. But no... the certificate error was showing on the Google search page and I hadn't clicked through at all.. so why was Google trying to load the page and showing the HTTPS error because of the invalid certificate?
-----------.com:443 uses an invalid security certificate.
The certificate is not trusted because it is self signed.
(Error code: sec_error_untrusted_issuer)
It could be a problem with the server's configuration or it could be someone trying to impersonate the server.
If you have connected to this server successfully in the past the error may be temporary and you can try again later.
The answer lies in prefetch - a combination of a tag on the site, Google and the default browser configuration meant that the browser tried to automatically load content from the bad site just by Googling for something.
Link prefetching (and how to turn it off) is explained in this FAQ or this HOWTO guide.. if you are using a Mozilla based browser then go and turn if off NOW by going into about:config and setting network.prefetch-next to false.
So why is it so dangerous? Have there been any cases of malware using link prefetching to spread? Not as I know.. although it might be theoretically possible. The danger is that you have just revealed your IP address without knowing it..
Let's look at a particular scenario where this can be used. Let's say the attacker is targetting a victim who is using an unidentifiable email address, and the attacker wants to find that victim's IP to tie them down to a location or organisation. In this scenario, the victim is not stupid.. they don't click on links in spam, they don't reply to untrusted messages, never send read receipts and they don't load external images in their mail client.. but the attacker uses social engineering to send an email with details that the victim might Google (for example a telephone number). The victim may then search for references on Google and even without clicking on anything, the prefetch may reveal their IP address.
Alternatively, prefetch could be used to download illegal content onto a target machine without the victim knowing about it, or there are probably several other ways in which it can be abused.
So it's hard to tell if the original spam was a Joe Job, or someone using prefetch to collect IP addresses for evil purposes. But I'll bloody well keep the prefetch switched off in future..