I wonder if it's related: for the last weeks I'm getting "you've been blocked" screens when visiting random websites. Is there a way to know if my IP made it to one of those blacklists?
There's decent literature on proxy/vpn detection by network and timing analysis. But I have the impression it's based on the inspection an ISP or MitM is able to do, not really the receiver of a request AFTER it goes out of the proxy node.
agree with the ending part of the post: it's important to avoid shared IPs causing wrong classifications, but even more important you can't really block people for using proxies. What matters is the big picture at the site traffic's level
that may be true in some cases, but I run a small site that gives free credits to new accounts and I get lots of fake and abusing accounts created with the VPN + fake email combo. I know because when I look up the IP it's always a datacenter or hosting range.
I wonder if it's related: for the last weeks I'm getting "you've been blocked" screens when visiting random websites. Is there a way to know if my IP made it to one of those blacklists?
you can check you own IP at abuseipdb.com
the demo is nice but I can't find details on how it does it internally?
they mention network analysis, packet timings and format. Not sure how reliable that is
There's decent literature on proxy/vpn detection by network and timing analysis. But I have the impression it's based on the inspection an ISP or MitM is able to do, not really the receiver of a request AFTER it goes out of the proxy node.
agree with the ending part of the post: it's important to avoid shared IPs causing wrong classifications, but even more important you can't really block people for using proxies. What matters is the big picture at the site traffic's level
that may be true in some cases, but I run a small site that gives free credits to new accounts and I get lots of fake and abusing accounts created with the VPN + fake email combo. I know because when I look up the IP it's always a datacenter or hosting range.