# LAN DNS

The homelab gives every service a name under `franksdank.cloud`. Users type a name; they never type a port number. This page covers how names reach users: the Pi-hole server, the G36 router hand-off, and the split between LAN and WAN resolution.

See [[Caddy]] for the reverse proxy that serves the names, and [[Network Architecture]] for the full topology.

## Components

| Component | Where | Role |
|-----------|-------|------|
| Pi-hole | Dankclaw quadlet (`dankclaw/pihole/`) | Answers the flat service names for LAN clients |
| SURFboard G36 | 192.168.0.1 | Hands dnsmasq to clients as the DHCP DNS server |
| Porkbun | Registrar for `franksdank.cloud` | Answers the same names for the WAN |
| Caddy | Dankclaw (192.168.0.143:443) | Serves every name over HTTPS |

## How a name reaches a client

```mermaid
sequenceDiagram
    participant C as LAN client (laptop, phone)
    participant G as G36 router (192.168.0.1)
    participant D as Pi-hole (192.168.0.143:53)
    participant P as Porkbun (WAN)
    participant X as Caddy (192.168.0.143:443)

    C->>G: DHCP request
    G-->>C: lease + DNS = 192.168.0.143
    C->>D: dig romm.franksdank.cloud
    D-->>C: 192.168.0.143
    Note over D: addn-hosts answer<br/>(dankclaw/pihole/addn-hosts)
    C->>X: https://romm.franksdank.cloud
    X-->>C: RomM (192.168.0.235:8085 upstream)
    C->>P: dig franksdank.cloud (unknown name)
    P-->>C: 71.227.227.172
```

Pi-hole answers only the names in `addn-hosts`. Every other query goes to the upstream resolvers `1.1.1.1` and `1.0.0.1`. The zone apex and unknown subdomains fall through to Porkbun. This avoids a wildcard override.

## Why split DNS

The G36 has no NAT loopback (hairpin). A LAN client that resolves a name to the WAN IP and then connects to it would time out. Split DNS removes the hairpin: LAN clients get the LAN IP, WAN clients get the public IP. Both use the same name and the same Let's Encrypt certificate. See [[Caddy]] for the TLS side.

## Pi-hole quadlet

Pi-hole FTL is a dnsmasq fork. It serves the same `addn-hosts` file the retired dnsmasq quadlet used, mounted into `/etc/pihole/hosts/` read-only. It retired the dnsmasq quadlet on 2026-08-24.

- **Location:** `dankclaw/pihole/` — `quadlet/pihole.container`, `env.example`, `addn-hosts`
- **Ports:** 53/udp + 53/tcp (DNS) and 8083/tcp (web UI) on 192.168.0.143
- **Host prerequisite 1:** `sysctl net.ipv4.ip_unprivileged_port_start=53` (persisted in `/etc/sysctl.d/99-dnsmasq.conf`)
- **Host prerequisite 2:** `DNSStubListener=no` in `/etc/systemd/resolved.conf` (the resolved stub would otherwise hold port 53 inside the container netns)
- **Env file:** `/home/dank/Containers/pihole/env` (never committed) — `ServerIP`, `TZ`, `DNS1` + `DNS2` (pinned upstreams `1.1.1.1` + `1.0.0.1`), `WEBPASSWORD` (the shared family password)
- **Web UI:** `pihole.franksdank.cloud` (Caddy vhost → 8083), family login
- **Deploy + restart after a change:** `./scripts/deploy-pihole-quadlet.sh dankclaw`

Add a new service name in two places:

1. `dankclaw/pihole/addn-hosts` — `192.168.0.143 <name>.franksdank.cloud`
2. Porkbun — A record `<name> 71.227.227.172` (so WAN clients and the DNS-01 challenge work)

Then add the vhost to [[Caddy]] and run `./scripts/deploy.sh dankclaw`. See [[Pi-hole]] for the full configuration.

## G36 router DNS hand-off

The router must tell DHCP clients to use Pi-hole. Until that step is done, new clients get the router's own DNS relay and the flat names do not resolve for them. Existing clients that already use Pi-hole are unaffected.

**Manual step (web GUI at `https://192.168.0.1`):**

1. Open **Gateway → Connection → Local IP Network → IPv4**.
2. Uncheck **DNS Proxy** (disable the router DNS relay).
3. Set **DNS** to **Statically configured**.
4. Set **Primary DNS** to `192.168.0.143`.
5. Leave the secondary field empty or set `0.0.0.0`.
6. Apply, then re-lease DHCP on each client (`sudo dhclient -r && sudo dhclient`).

The router admin interface has no programmatic API in this firmware. The web GUI is the only control path. Keep the credentials out of this repo; the repo tracks no router secrets.

Verify after the change:

```
# on a LAN client
resolv.conf          # nameserver 192.168.0.143
dig romm.franksdank.cloud +short
# expected: 192.168.0.143
```

## Dankclaw host DNS

The Dankclaw host itself resolves the flat names through Pi-hole:

- `/etc/systemd/resolved.conf`: `DNS=192.168.0.143 192.168.0.1` (Pi-hole first, router as fallback)

Without this, the host would resolve the names through the router to the WAN IP and time out (no hairpin).

## Dankware host DNS

`dankware` is a LAN client, not a Pi-hole server. Its flat names resolve through Pi-hole (`192.168.0.143`), but its resolver setup differs from Dankclaw's because it carries two `+DefaultRoute` links.

- **Two default links:** `enp47s0` (ethernet, metric 100) and `wlan0` (wifi, metric 600), each getting the same DHCP DNS list `[192.168.0.143, 1.1.1.1, …]`.
- **The flap:** `systemd-resolved` picks a different "current" resolver per link — `enp47s0` settled on the Comcast v6 upstream (→ WAN IP, times out on the LAN), `wlan0` on Pi-hole (→ LAN IP). Flat-name resolution flapped between the two on every query.
- **The fix:** pin `192.168.0.143` as the static IPv4 resolver on both NetworkManager connections. `resolved.conf`'s own `DNS=` is a per-link fallback, so the link itself must be pinned:

```bash
for c in "Wired connection 1" FRANKFI-5G; do
  nmcli connection modify "$c" ipv4.method auto \
    ipv4.dns 192.168.0.143 \
    ipv4.ignore-auto-dns yes \
    ipv4.ignore-auto-routes no
done
nmcli connection up "Wired connection 1"
nmcli connection up FRANKFI-5G
systemctl restart systemd-resolved
```

`method:auto` keeps DHCP for the address/gateway/route (dual-default traffic path preserved); `ignore-auto-dns=yes` uses only the static resolver; `ignore-auto-routes=no` keeps the DHCP routes. Verify with a cache-clear loop — `romm.franksdank.cloud` should return `192.168.0.143` on every run:

```bash
for i in $(seq 1 10); do resolvectl flush-caches; dig romm.franksdank.cloud +short; done
```

The apex and unknown subdomains still fall through to Porkbun (→ `71.227.227.172`), so WAN clients and a bare `dig franksdank.cloud` keep working.

## WAN records (Porkbun)

The Porkbun zone `franksdank.cloud` carries an A record for every vhost plus the apex, all pointing at `71.227.227.172`. TTL is 300. The zone has no wildcard record: unknown subdomains return NXDOMAIN. Caddy's DNS-01 challenges add and remove `_acme-challenge.*` TXT records through the same API during issuance. See [[Caddy]] for the TLS flow.

`franksdank.casa` is not in this Porkbun account. The `plex.franksdank.casa` vhost stays plain HTTP for that reason. The canonical Plex name is `plex.franksdank.cloud` (HTTPS).

## Failure modes

| Symptom | First check |
|---------|-------------|
| Name does not resolve on a new client | `resolv.conf` shows 192.168.0.1, not .143 — the G36 hand-off step is missing |
| Name resolves to 71.227.x and times out on the LAN | Client used the router DNS — same root cause |
| Flat name flaps between 192.168.0.143 and 71.227.x on one client | Two `+DefaultRoute` links; resolved picks a different "current" resolver per link. Pin 192.168.0.143 as static DNS on both NM connections — see [[#Dankware host DNS]] |
| `dig @192.168.0.143` fails entirely | Pi-hole quadlet down: `systemctl --user status pihole` |
| Single-label name (for example `dankclaw`) answers `127.0.1.1` | The rootless container `/etc/hosts` carries the host entry (`127.0.1.1 <host>`). FTL serves it. Use the FQDN, or an `/etc/hosts` entry on the client |
| HTTPS works from the host but not the WAN | No port forward on the G36 for that service (manual router step) |
| Cert renewal fails in Caddy logs | Porkbun API credentials in the Dankclaw `.env`; test with the Porkbun API `/ping` endpoint |

## See Also

- [[Pi-hole]] — the LAN DNS server (quadlet, records, web UI)
- [[Caddy]] — TLS, vhost list, DNS-01
- [[Network Architecture]] — topology, access policy
- [[Dankclaw Stack]] — the host that carries Pi-hole and Caddy