AuthHound

Ignoring request to auth address ... from unknown client

FreeRADIUS 'unknown client' — why the NAS times out, not fails

The log line:

Ignoring request to auth address * port 1812 bound to server default
  from unknown client 192.168.10.99 port 49152 proto udp

What it actually means: the packet reached FreeRADIUS, but no client {} block in clients.conf covers its source IP — so the server dropped it without any reply. That silence is the defining symptom: the NAS doesn’t get a reject, it gets nothing, retries, and eventually reports a timeout or “RADIUS server not responding.” Users just see Wi-Fi that spins forever.

If you’re debugging “the RADIUS server isn’t responding” from the NAS side, this log line is the first thing to look for — the server is responding fine to everyone it knows; it’s deliberately ignoring this one address.

The three causes

1. There’s simply no client entry yet

New access point, new switch, or a NAS whose management IP changed. Add a block with the exact source IP from the log line (not the IP you think the NAS has):

client wlc-02 {
    ipaddr = 192.168.10.99
    secret = <shared secret>
}

Then restart radiusd (or send SIGHUP) — clients.conf is read at startup, not per-request.

2. The NAS sends from a different interface than you whitelisted

Multi-homed controllers and switches pick a source interface by routing table, which is often the uplink IP rather than the management IP you configured. NAT between NAS and server does the same thing wholesale.

Check: compare the source IP in the log line with the address in your client entry. On the NAS, look for a “RADIUS source interface” setting and pin it to the management interface — that’s the durable fix; whitelisting whatever IP happens to show up is the fragile one.

3. The entry exists but isn’t loaded

Config never reloaded after the edit, the entry sits in a file that isn’t included, or it belongs to a virtual server that isn’t listening on this address.

Check: radiusd -X prints the full client list at startup — search it for your IP. If it’s missing there, the config on disk isn’t what the server is running.

Note for wide networks

If you have many NAS devices on one subnet, a range entry (ipaddr = 192.168.10.0/24) beats twenty host entries — but give the whole range one secret and never let a broader range overlap a host entry with a different secret, or you’ll trade this error for shared-secret mismatches that are far harder to spot.

Need the clients.conf block this error is asking for? Generate it — runs in your browser →

Test it from a client’s position in one command: the open-source authhound-probe sends a real Access-Request from inside your network and tells you whether the server is unreachable, not listening, or silently ignoring you. Comma-separate two servers to compare them; add --count 10 if the failure comes and goes.

export AUTHHOUND_RADIUS_SECRET='shared-secret'   # not in argv or shell history
authhound-probe radius test --server radius.corp.com --peap alice

Quickstart: register the probe as a client and run the first test

Have the full log open? Paste it here

The analyzer detects this exact failure and 18 others, and ranks the likely causes for your specific output. It runs entirely in your browser — nothing is uploaded.

This failure is usually invisible until users complain. AuthHound is building continuous monitoring that catches it first — early access:

No spam, no sharing — one email when it's ready, and you can reply to be removed anytime.