Panic over "Incoming DTLS data is not being consumed"

[zRedShift]

I'm opening this issue to track a rare crash we've had on our SFU server using str0m.
We're also using TCP with RFC4571 framing as a fallback, and I suspect this might cause big messages, but it's also possible that something is wrong with openssl and it occurred under regular UDP.
I'm gonna add some logs and hope this reproduces.

str0m/src/crypto/ossl/io_buf.rs

Lines 18 to 21 in 95ebe52

	assert!(
	self.incoming.len() < 30_000,
	"Incoming DTLS data is not being consumed"
	);

The application panicked (crashed).
Message:  Incoming DTLS data is not being consumed
Location: /root/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/str0m-0.9.0/src/crypto/ossl/io_buf.rs:18

Backtrace omitted. Run with RUST_BACKTRACE=1 environment variable to display it.
Run with RUST_BACKTRACE=full to include source snippets.