pszabop

Ditto here. I need to be able to call async logging functions, async requests out to external processes, etc. No problem with `request_body_filter`, it's already asynchronous. This appears to be...

interesting approach. I'm working on using the `-u` upgrade feature to change the config (or the binary), rather than trying to keep a lot of dynamic config around in pingora....

If pingora upgrade is seamless, then unstable Pod IPs are not a problem. Upgrade every few minutes. But upgrade may not be as seamless as one would hope for (e.g....

Yep, that did it. So this is a documentation request or a "pick suitable defaults for responses when request_filter returns Ok(true)"

note I haven't implemented the load balancing algorithm type yet, which will add a whole new layer of complexity since the load balancer takes the algorithm as a type instead...

and, didn't implement background health monitor, the above model is completely broken.

Because of some weird ownership issues on LoadBalancers (which own the Backends), where the LoadBalancer is owned by the BackgroundService, you can't easily have two BackgroundServices referring to the same...

Patch to `rc/protocols/tls/boringssl_openssl/stream.rs` ``` + use crate::protocols::tls::boringssl_openssl::stream::ssl::NameType; + let sni = ssl.servername(NameType::HOST_NAME); + let sni_string: Option = sni.map(ToOwned::to_owned); SslDigest { cipher, version: ssl.version_str(), organization: org, serial_number: sn, cert_digest, + sni:...

I removed the patch to Dockerfile, it's clearly out of date with the github build actions and thus should be handled as a separate issue. Now handling local builds with...

the entire SSL Digest creation method was changed and it causes merge conflicts and the function signature doesn't take the SNI. So will have to go through some effort to...