How do I block troublesome IP ranges?

What subnets?

[Finder]

Chinese and Korean IPs make up the vast majority of whois lookups I do when I catch someone snooping around, making bad requests, or otherwise wasting bandwidth.

If I wanted to block the worst offenders, what IP subnets would I be looking at?

Has anyone else done this, and if so, what were your results?

I run a couple of informational sites, but they are very niche-specific and I've never seen valid traffic from an Asian IP.

[seindal]

I block email from china and korea because of spam. For that I use to DNS based blacklists.

The lists are at: korea.services.net and cn.rbl.cluecentral.net

They work like this.

If you have a request from, say, IP 208.60.161.180 (not korea), you query the name-server for 180.161.60.208.korea.services.net (note how the numbers are reversed). In this case no data are returned, so the IP is not in Korea. If we try with IP 211.244.251.100, the DNS lookup for 100.251.244.211.korea.services.net returns a record and the IP is in Korea.

Likewise for the China list.

Obviously, the people behind these lists have the actual IP blocks in question, but I don't know if they will give them away. The lists might not be static. That is the force of have a system based on dynamic lookup.

Anyway, this might be a point of departure for finding the IP-blocks.

René.

[andreasfriedrich]

Funny thing that people from the west are thinking about blocking access from China while the Chinese government tries to block access to the outside ;)

Of course the first is perfectly legal while the latter is not.

[seindal]

There is an important difference, in that I personally decide that I don't want email from Korea and China because of the signal/noise ratio of the communication from there. There is no goverment imposing this on me.

I have never ever received a legitimate mail from any of the two countries. I do know somebody from Korea, but they use hotmail, probably because mail from there Korean addresses consistently bounces :)

René.