Sometimes either IPv6 or IPv4 connection is fucked up, not on hoster's side. This is a today's example:
laura@hakase:~$ traceroute graph.facebook.com -4
traceroute to graph.facebook.com (31.13.81.9), 30 hops max, 60 byte packets
1 edge01.waw.bgp.wtf (185.236.240.1) 1.028 ms 1.092 ms 1.013 ms
2 facebook.ip4.epix.net.pl (89.46.145.202) 26.891 ms 26.915 ms 26.898 ms
3 * * *
4 * * *
5 * * *
6 * * *
7 * * *
[nothing happens later, just packet loss]
This yet another disaster at EPIX obviously breaks the mautrix-facebook bridge. Of course, one could say that in this situation you should tell your provider to scream at their provider, re-route the traffic to avoid the broken provider, and so on, and they'll be absolutely right! But anyway, I think this kind of situation should also have a quick workaround - an option in the bridge config, like one of these:
prefer_ip_version: 6
prefer_address: inet6
prefer_ipv6: true
Sometimes either IPv6 or IPv4 connection is fucked up, not on hoster's side. This is a today's example:
This yet another disaster at EPIX obviously breaks the mautrix-facebook bridge. Of course, one could say that in this situation you should tell your provider to scream at their provider, re-route the traffic to avoid the broken provider, and so on, and they'll be absolutely right! But anyway, I think this kind of situation should also have a quick workaround - an option in the bridge config, like one of these: