Did you read the post? Here's just one example: Through packet level filtering at the firewall it’s possible to apply rules to an entire shared server, b

Did you read the post? Here's just one example: Through packet level filtering at the firewall it’s possible to apply rules to an entire shared server, blocking the abuse immediately. For example, let’s say someone decides to use TorGuard to unlawfully promote their Ugg boots business (spam). In order for us to block this one individual, we simply implement new firewall rules, effectively blocking the abused protocol for everyone on that VPN server. Since there are no user logs to go by, we handle abuse per server, not per user. Seems like a silly way to handle it since eventually none of their servers will be able to access the internet.