ORPort incorrectly detected as non-reachable in 0.4.5.6/6

Summary

After my bridge upgrade from 0.4.3.5 to 0.4.5.5 (and now 0.4.5.6 - from your "debian stable main" repository) my ORPort isn't detected as reacheable anymore (a message in logs after 20 minutes). However, when I check it with telnet or even using your TCP reachability test it is completely fine. I suspect, it is a regression related to more aggressive IPv6 mode (I've seen the other issues, but they seemed to be fixed in 0.4.5.6).

I tried to set ORPort to "55555 IPv4Only", but it does seem to be ignored. I still observe:

Unable to find IPv4 address for ORPort 55555. You might want to specify IPv4Only to it or set an explicit address or set Address.

maybe related to #40300 (closed).

As I run Tor in a container image in Podman, I cannot easily define an IP address in the configuration. In addition the container might support IPv6 at its network interface, but my external (host's) network does not. That might be confusing for Tor and maybe the external checker tries to reach it using IPv6.

It would be good to be able to simply disable IPv6 globally (IPv4Only in ORPort doesn't work for me). Related to #40208 (closed)? Is there any workaround available right now to make it work again?

Environment

Tor 0.4.5.6 from your "debian stable main" repository

Tor 0.4.5.6 running on Linux with Libevent 2.1.8-stable, OpenSSL 1.1.1d, Zlib 1.2.11, Liblzma 5.2.4, Libzstd 1.3.8 and Glibc 2.28 as libc.