Joe Birr-Pixton

This is sufficient to reproduce: ```patch diff --git a/rustls/tests/unbuffered.rs b/rustls/tests/unbuffered.rs index 97a0f189..eccc04b6 100644 --- a/rustls/tests/unbuffered.rs +++ b/rustls/tests/unbuffered.rs @@ -100,7 +100,8 @@ fn tls13_handshake() { } fn handshake(version: &'static rustls::SupportedProtocolVersion) ->...

> Thanks @ctz . I tried to set the `max_fragment_size` but I still cannot reproduce the issue. My understanding is decreasing the fragment size results in smaller TLS messages, thereby...

> Might be interesting to see if you can isolate this in a test binary directly using rustls-platform-verifier. (on a platform other than linux :) )

Likely we won't support this in the core crate's providers for quite some time, as it's not well motivated by a reasonable threat model. Here's some examples of events that...

This is unlikely to be related to the server certificate. Instead, the server does not have any overlap with the client's TLS settings, so it sends a TLS alert saying...

> I'm facing the same (similar?) issue. I've heavily redacted the following output to spare the ~guilty~ innocent. I'm not SSL expert but I think the `sslscan` output is telling...

I have put this back to draft. Please work on making this reviewable, by (for example) minimising back-and-forth changes, making a sequence of logical commits, and perhaps slicing off some...

Yes that seems like a typo in Cargo.toml. In practice it doesn't have any effect (because every optional dependency implies a feature of the same name).

> > These now come from IANA, with significant filtering. > > Out of curiosity, was this manual or programmatic? I pushed the script here: https://github.com/ctz/junk-drawer/blob/main/iana-csv-scripts/iana.py (but I view it...

Will bring this back sometime later.