You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In Fedora 35, and possibly other distributions using systemd-resolved for DNS resolution, there are times when reverse DNS resolution seems to hang and never complete in bzfs. The specific instance I found was a case where there either wasn't a reverse DNS hostname, or the server responsible for providing the reverse DNS wasn't responding. After installing the bind-utils package to acquire the host command, I attempted to resolve the reverse DNS of the player's IP. I received a "connection timed out; no servers could be reached" message after several seconds. When I ran the same command but also passed a DNS server to use, I instead received a SERVFAIL almost immediately.
It's possible we're not handling some result from c-ares where a timeout occurs.
The text was updated successfully, but these errors were encountered:
This occurred for me with bzfs 2.4.24 and systemd-resolved 251.7 on NixOS. The symptom I observed was that when I connected to bzfs from my LAN, the client hung on the "Press Right Mouse or i to join".
Two independent workarounds:
Set nameserver in /etc/resolv.conf to a public resolver.
Set up a port redirection of 5154/{udp,tcp} on your server using socat (for example) so that bzflag sees the connections as coming from localhost. This avoids triggering the same issue.
In Fedora 35, and possibly other distributions using systemd-resolved for DNS resolution, there are times when reverse DNS resolution seems to hang and never complete in bzfs. The specific instance I found was a case where there either wasn't a reverse DNS hostname, or the server responsible for providing the reverse DNS wasn't responding. After installing the bind-utils package to acquire the host command, I attempted to resolve the reverse DNS of the player's IP. I received a "connection timed out; no servers could be reached" message after several seconds. When I ran the same command but also passed a DNS server to use, I instead received a SERVFAIL almost immediately.
It's possible we're not handling some result from c-ares where a timeout occurs.
The text was updated successfully, but these errors were encountered: