- Newest
- Most votes
- Most comments
One of the purposes of CloudFront is to protect your application from attacks - trying to create an architecture where you're trying to protect CloudFront is going to create a lot of complexity and cost for you - assuming that you can get it to work anyway.
Given that you are using AWS WAF, why not use that in conjunction with CloudFront to deal with attacks that you are concerned about?
To your actual question: CloudFront only accepts HTTP and HTTPS connections from clients. It does not support PPv2 and you cannot get WAF on CloudFront to interpret PPv2 for you because CloudFront has to first terminate the HTTP/HTTPS connection and then pass the session onto WAF. If the session it receives is not valid HTTP/HTTPS it will not accept it - and that excludes a session which is using a PPv2 header.
Relevant content
asked 2 years ago
- AWS OFFICIALUpdated 9 months ago

Hello,
we'll stop using AWS WAF, because Radware is (theoretically) smarter and easier to maintain, especially from a logging perspective. Clear, thanks for the response, so no matter what we put in front of the CDN there's no way to manipulate the request into letting the CDN think it's comining from the actual client. Good to know. In the meantime we started exploring the Lambda integration of Radware, that might be the better solution for us, if we're sure it works like we think (with reference to my question about Lambda optimisations). Thanks again.