database64128

A new release is out with these new features. Closing this issue.

> 1. Did you really fill the 2,000,000,000 µI conduit buffer? > 2. Does the induction matrix accept input on the output connection, too? > 3. Did you do anything...

/cc @laktyushin

This does not appear to work. Renovate still ignores the `toolchain` line and forcibly bumps the `go` line: https://github.com/database64128/tfo-go/commit/4074a20d745faeff121a45c66ec140918e282e5a (https://github.com/database64128/renovate-config/blob/812dbffc3e53a061e42f47c51d6de2e8f3ef14ed/go.json). #16715 should probably be reopened as well.

@rarkins Opened #28468

It seems like an MTU misconfiguration somewhere in your network. You didn't discover the issue with shadowsocks-libev because it didn't bother with disabling IP fragmentation.

This indeed would be something really nice (and important IMO) to have in shadowsocks-rust. My project [shadowsocks-go](https://github.com/database64128/shadowsocks-go) was written with this in mind and has a mandatory MTU option for...

> So in practice, the `mtu` value in configuration, will limits the buffer size that used to `recv()` from the UDP socket, which is `mtu - tag_size - header_size` in...

> Ok. But what should application do to handle these oversized packets? Just ignore them? My take is to print a warning and drop the packet. > Theoretically, the `mtu`...

Another reason to implement the MTU option: QUIC. Many QUIC clients use DPLPMTUD ([RFC 8899](https://www.ietf.org/rfc/rfc8899.html)) to probe the PMTU and determine the best packet size. If the proxy program allows...