https://dustri.org/b/grapheneoss-mte-caught-a-crash-in-signal.html
The issue never made its way to production, and only happened during 7.10.0-beta: it was introduced in 11557e4815e5acc429db79945b94c9e5465eed77 the 24th of June 2024, reported to Signal's security team likely reported the 26th of June 2024, fixed in 37815a3f39661b4331c4ed26fbd640618e4b7147 the very same day. It's interesting to notice that the fix is a couple of lines of safe-looking pure Java code, and not some C++, making it hard to spot.
Anyway, it's always nice to see this kind of memory-corruption issues being caught "in the wild".