Web interface has mysteriously "disappeared"

A few months back, I set up a TrueNAS CORE system. It presently reports its version as TrueNAS-13.0-U6.2 (6bf2413add).

More recently, a client stopped being able to complete a Time Machine backup. There isn’t enough space on the server, says the client. Fine. That’s a thing that happens.

So I think: It has come time to delete some old periodic ZFS snapshots on the server side. Or maybe I’ll even configure the server to purge old snapshots automatically or something. Details of that don’t matter yet.

I choose my bookmark for the server and… whoops. Web browser says server isn’t responding. More specifically, Firefox say it “can’t establish a connection.”

So I think: That can’t be right. I know the server is up generally and I know haven’t touched it or the network in forever.

So I try to SSH in and sure enough that’s possible. I see nginx processes running, but apparently they aren’t responding to the outside world. Superstitiously, I try another web browser, which gives me a similarly sad response.

Via SSH, I superstitiously reboot the server and wait. Stuff (SMB, SSH, a jail on a distinct IP address) seems to come back up generally, but the web GUI remains inaccessible. Same error. So: No change in behavior that I have detected.

Any ideas?

Obviously this has not been an emergency for me.

In particular, Time Machine clients were able to resume backups. Don’t know why it stopped working! Don’t know why it started working again! Mysterious! Fun!

But just now I thought to check if web access had also resumed, and no, it has not. Everything seems the same, including my ability to SSH in and observe:

# ps ax | grep nginx
 1585  -  IWs       0:00.00 nginx: master process /usr/local/sbin/nginx
 1586  -  IW        0:00.00 nginx: worker process (nginx)

I suppose the fact that they have both gotten no execution time is consistent with web access being down. But I don’t know how to diagnose more deeply than that.

Ideas?