Help regaining a CLI

For a while now, I've been unable to use SSH because of "too many open sockets."

# IPv6 addresses slightly edited

Jul 21 05:30:36 zx3 rc.sshd: Starting SSH server daemon...
Jul 21 05:30:36 zx3 rc.sshd: /usr/sbin/sshd
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 2001:db8:b9dd:900::d port 22.
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 10.9.0.13 port 22.
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 2001:db8:b9dd:800::d port 22.
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 10.8.0.13 port 22.
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 2001:db8:b9dd:700::d port 22.
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 10.7.0.13 port 22.
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 2001:db8:b9dd:600::d port 22.
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 10.6.0.13 port 22.
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 2001:db8:b9dd:1200::d port 22.
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 10.18.0.13 port 22.
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 2001:db8:b9dd:1100::d port 22.
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 10.17.0.13 port 22.
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 2001:db8:b9dd:1000::d port 22.
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 10.16.0.13 port 22.
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 2001:db8:b9dd:e00::d port 22.
Jul 21 05:30:36 zx3 sshd[40808]: Server listening on 10.14.0.13 port 22.
Jul 21 05:30:36 zx3 sshd[40808]: fatal: Too many listen sockets. Enlarge MAX_LISTEN_SOCKS
Jul 21 05:30:36 zx3 rc.sshd: SSH server daemon...  Failed.
Jul 21 05:30:36 zx3 emhttpd: shcmd (2083): rm -f /boot/config/plugins/dynamix/mover.cron
Jul 21 05:30:36 zx3 emhttpd: shcmd (2084): /usr/local/sbin/update_cron
Jul 21 05:30:36 zx3 emhttpd: Starting services...

I still could use the web shell it has most of the time, sometimes it would just open a black window that would never print any text in it other than a blinking cursor. I never found what triggers its failure.

In the last few days, the shell with the prompt that never gets ready became the norm effectively locking me out of all CLI but the console itself. The server room, a former walk-in closet, is both freezing and burning, no desk but only the cold floor tile to sit on. It's extremely uncomfortable for more than a quick check. Never mind console use.

I read somewhere that it may be do to changes in the go file, and I had edited it to change the default contents of /root but since then I undid the changes but I'm still locked away. I remember also verifying ~/.ssh permissions to be correct, but I don't remember where if the directory was a symlink or not. In any case, that should be irrelevant for the web CLI, right?

In some xBSD forum I stumbled on that the solution for this could potentially be making sshd listen in 0.0.0.0 and ::1 rather than on every interface; then use FreeBSD's firewall to police access. Aside that there's a few missing key pieces that solution in Unraid, namely the firewall, as far as I'm aware, /etc is expanded from the Republic of Congo or whatever straight into memory each boot, rewriting whatever was held live since the last boot. Though since it was never there in the first place, maybe rewriting is not the best term.

Any ideas?

Thanks.