docs: re-home ACME client walkthroughs under docs/migration/

The three ACME client walkthroughs (Caddy, cert-manager, Traefik) are
conceptually "I have an existing X, here's how to point its ACME
client at certctl." They belong with the migration docs, not with the
acme-server protocol reference.

Renames:
  docs/acme-caddy-walkthrough.md       → docs/migration/acme-from-caddy.md
  docs/acme-cert-manager-walkthrough.md → docs/migration/acme-from-cert-manager.md
  docs/acme-traefik-walkthrough.md     → docs/migration/acme-from-traefik.md

Each walkthrough's lede gets a "Use this walkthrough when..." paragraph
that closes the WHY-weak gap flagged in the Phase 1 audit. The new
framing tells the reader when to pick this walkthrough versus the
alternatives:

  - Caddy: "you're running Caddy 2.7+ and want it to ACME-issue from
    certctl instead of Let's Encrypt"
  - cert-manager: explicit pointer to cert-manager-coexistence.md for
    the keep-cert-manager-running case (vs replacement)
  - Traefik: "you're running Traefik 3.0+ and want certctl as your
    ACME source of truth"

Cross-reference updates from other docs and README still pending in
Phase 11.
This commit is contained in:
shankar0123
2026-05-05 02:51:10 +00:00
parent 32009cf7c8
commit 2e8c5a8d22
3 changed files with 23 additions and 0 deletions
+179
View File
@@ -0,0 +1,179 @@
# Caddy Integration Walkthrough
> **Use this walkthrough when** you're already running Caddy 2.7+ and
> want it to ACME-issue from certctl (your internal CA, your private
> PKI, or a local sub-CA chained under an enterprise root) instead of
> Let's Encrypt. The Caddyfile changes are minimal; the load-bearing
> piece is trusting certctl's bootstrap CA so Caddy's ACME client can
> talk to certctl over HTTPS.
End-to-end recipe for issuing certs from a certctl-server deployment
through Caddy 2.7+. Target audience: operator running Caddy on a VM
or container who wants Caddy to ACME-issue from certctl instead of
Let's Encrypt.
## Prereqs
- A reachable certctl-server with `CERTCTL_ACME_SERVER_ENABLED=true`
and at least one profile whose `acme_auth_mode` is set. Profile
setup is identical to the cert-manager walkthrough — see
[`docs/acme-cert-manager-walkthrough.md`](./acme-cert-manager-walkthrough.md)
Step 2.
- Caddy 2.7.x or later. `caddy version` should show 2.7.0+.
- Network reachability: Caddy → certctl-server's HTTPS listener (port
8443 by default).
- The certctl bootstrap CA, in PEM form, captured for the trust
configuration below. Capture exactly the same way as the cert-manager
walkthrough Step 3 — use `cat deploy/test/certs/ca.crt`.
## Step 1 — Configure Caddy
Caddy's ACME issuer is configured per-site (or globally) via the
`acme_ca` directive in a Caddyfile, or via the `tls.acme_ca` field
in JSON config. The directive points at the directory URL:
```
{
email ops@example.com
}
example.com {
tls {
acme_ca https://certctl.example.com:8443/acme/profile/prof-test/directory
issuer acme
}
reverse_proxy localhost:8080
}
```
Notes:
- `acme_ca` must point at the directory URL (ending in `/directory`),
not just the base. Caddy uses the directory document to discover
the new-account / new-order URLs, exactly the same way cert-manager
does.
- `issuer acme` is the default; included here for clarity. Caddy can
also be configured with `issuer zerossl` or `issuer internal`; for
certctl integration, `acme` is the correct issuer.
- Caddy auto-discovers `tls-alpn-01` first when port 443 is bound to
Caddy, then falls back to HTTP-01. For `trust_authenticated` mode
profiles, both work without solver round-trips.
## Step 2 — Trust the certctl bootstrap CA
Caddy validates the certctl-server's TLS chain before any ACME call,
the same way cert-manager does. Two options for trust:
### Option A — OS trust store (preferred for VMs)
```
sudo cp deploy/test/certs/ca.crt /usr/local/share/ca-certificates/certctl-bootstrap.crt
sudo update-ca-certificates
sudo systemctl restart caddy
```
Caddy honors the system trust store via the Go runtime's
`crypto/x509` defaults. After `update-ca-certificates`, Caddy's HTTPS
client trusts certctl's self-signed root and the directory call
succeeds.
### Option B — Caddy `tls.cas` (for containerized deployments)
```
{
pki {
ca certctl_bootstrap {
root_cert_file /etc/caddy/certctl-bootstrap.crt
}
}
}
example.com {
tls {
acme_ca https://certctl.example.com:8443/acme/profile/prof-test/directory
ca certctl_bootstrap
issuer acme
}
reverse_proxy localhost:8080
}
```
The `pki.ca` block registers a named CA Caddy can reference; the
`tls.ca certctl_bootstrap` line in the site block scopes that trust
to ACME calls for this site only. This is the right pattern for
multi-tenant Caddy deployments where some sites trust certctl + others
don't.
## Step 3 — Reload Caddy
```
caddy validate --config /etc/caddy/Caddyfile
sudo systemctl reload caddy
```
Caddy reloads atomically; in-flight requests complete on the old
config while new requests use the new ACME issuer. On the next
`example.com` request, Caddy hits certctl's directory URL, registers
an account, submits a new-order, and finalizes — typically completing
in under 5 seconds for `trust_authenticated` mode.
## Step 4 — Verify
```
caddy list-certificates
# example.com (issuer=certctl.example.com): CN=example.com, valid until 2026-06-30
```
The cert is in Caddy's certificate cache (`$XDG_DATA_HOME/caddy/certificates/`
by default). Inspect:
```
openssl x509 -in ~/.local/share/caddy/certificates/acme-v02.api.letsencrypt.org-directory/example.com/example.com.crt -noout -subject -issuer -dates
# subject= CN=example.com
# issuer= CN=certctl test internal CA
```
(Path layout is Caddy-version-dependent; check `caddy environ` for the
canonical data dir.)
On the certctl side, the operator's audit log captures the issuance
event:
```
psql -c "SELECT actor, action, resource_id FROM audit_events
WHERE actor LIKE 'acme:%' ORDER BY created_at DESC LIMIT 5;"
```
## Common failure modes
- **Caddy logs `tls: failed to verify certificate: x509: certificate
signed by unknown authority`** → certctl bootstrap CA is not in
Caddy's trust path. Re-do Step 2; verify with `curl --cacert
/etc/caddy/certctl-bootstrap.crt https://certctl.example.com:8443/acme/profile/prof-test/directory`.
- **Caddy logs `urn:ietf:params:acme:error:rateLimited`** → certctl
per-account orders/hour limit hit (default 100/hr). Tune via
`CERTCTL_ACME_SERVER_RATE_LIMIT_ORDERS_PER_HOUR` if you have
legitimately high throughput.
- **Caddy logs `urn:ietf:params:acme:error:rejectedIdentifier`** →
the SAN list includes an identifier the certctl profile policy
rejects. Cross-reference [`docs/acme-server.md` § Troubleshooting](./acme-server.md#certificate-readyfalse-with-rejectedidentifier).
- **`badNonce` in Caddy logs** → clock skew or multi-replica certctl
without sticky sessions; same fix as the cert-manager walkthrough.
## Cleanup
```
caddy stop
# remove the certctl-specific block from your Caddyfile
sudo systemctl reload caddy
# Optional: delete cached certs from the certctl directory namespace.
rm -rf ~/.local/share/caddy/certificates/certctl.example.com-*
```
## See also
- [`docs/acme-server.md`](./acme-server.md) — canonical reference.
- [`docs/acme-cert-manager-walkthrough.md`](./acme-cert-manager-walkthrough.md) —
K8s-native equivalent.
- [Caddy upstream ACME docs](https://caddyserver.com/docs/automatic-https#acme-issuer)
— verify behavior pinned here against Caddy 2.7.x semantics.
+263
View File
@@ -0,0 +1,263 @@
# cert-manager Integration Walkthrough
> **Use this walkthrough when** you're already running cert-manager
> 1.15+ in Kubernetes and want it to issue certs from certctl (your
> internal CA, your private PKI, or a local sub-CA chained under an
> enterprise root) via the standard ACME `ClusterIssuer` model. If
> you want certctl to coexist with cert-manager rather than replace
> its issuer backend, see
> [`docs/migration/cert-manager-coexistence.md`](cert-manager-coexistence.md)
> instead.
End-to-end recipe for issuing certs from a certctl-server deployment
through cert-manager 1.15+. Target audience: Kubernetes operator who
has never deployed certctl before and wants a working
`Certificate``Secret` flow on their cluster in under 30 minutes.
The Phase 5 integration test (`make acme-cert-manager-test`) automates
exactly the recipe below. The YAML snippets in this doc are byte-equal
to the files under `deploy/test/acme-integration/` — re-running the
test from a fresh clone produces the same results documented here.
## Prereqs
- A Kubernetes cluster (kind / k3d / EKS / GKE / AKS / on-prem). For
local trial, `kind v0.20+` works exactly the way the Phase 5 test
uses it. The kind config lives at
[`deploy/test/acme-integration/kind-config.yaml`](../deploy/test/acme-integration/kind-config.yaml).
- `kubectl` v1.27+, `helm` v3.13+.
- `cert-manager` v1.15.0 installed in the `cert-manager` namespace.
If absent, run:
```
bash deploy/test/acme-integration/cert-manager-install.sh
```
which is the same idempotent installer the integration test uses.
- A certctl Helm chart published to a registry your cluster can pull
from. The Phase 5 test uses an `image.tag=test` placeholder; production
deployments use the actual image tag for your release line.
## Step 1 — Deploy certctl-server
```
helm install certctl-test deploy/helm/certctl/ \
--set acmeServer.enabled=true \
--set acmeServer.defaultProfileId=prof-test \
--set image.tag=test
kubectl wait --for=condition=Available --timeout=3m deployment/certctl-test
```
`acmeServer.enabled=true` flips the `CERTCTL_ACME_SERVER_ENABLED`
env var which gates the ACME route registration.
`acmeServer.defaultProfileId` sets `CERTCTL_ACME_SERVER_DEFAULT_PROFILE_ID`
so the `/acme/*` shorthand path mirrors the per-profile path family.
## Step 2 — Create the certctl profile
The ACME server requires a `certificate_profiles` row to bind issuance
to. Create one via the certctl API or GUI; for the simplest case set
`acme_auth_mode='trust_authenticated'`:
```
curl -X POST https://certctl-test.default.svc.cluster.local:8443/api/profiles \
-H 'Content-Type: application/json' \
-H "Authorization: Bearer $CERTCTL_API_KEY" \
-d '{
"id": "prof-test",
"name": "ACME test profile",
"issuer_id": "iss-internal-ca",
"max_ttl_seconds": 7776000,
"acme_auth_mode": "trust_authenticated"
}'
```
Auth-mode tradeoffs are covered in
[`docs/acme-server.md` § Auth-mode decision tree](./acme-server.md#auth-mode-decision-tree).
For first-time deployments, `trust_authenticated` is the right default.
## Step 3 — Capture the certctl bootstrap CA
cert-manager validates the certctl-server's TLS chain before sending
any account / order / finalize JWS. With certctl's self-signed
bootstrap cert (the demo default at `deploy/test/certs/server.crt`),
cert-manager rejects the directory URL with
`x509: certificate signed by unknown authority` unless you feed the
bootstrap CA in.
```
cat deploy/test/certs/ca.crt | base64 -w0
```
Capture the output for Step 4. This is **the** single biggest first-
time-deploy footgun on the cert-manager integration path. The reference
recipe lives in
[`docs/acme-server.md` § TLS trust bootstrap](./acme-server.md#tls-trust-bootstrap-read-this-before-configuring-cert-manager).
## Step 4 — Apply the ClusterIssuer
```yaml
# Phase 5 — sample ClusterIssuer for the certctl trust_authenticated
# auth mode (RFC 8555 §6 + certctl auth_mode=trust_authenticated, where
# the JWS-authenticated ACME account is trusted to issue any identifier
# the profile policy permits — no per-identifier ownership challenges).
#
# Use this as the starting template for any internal-PKI rollout.
# Replace the caBundle placeholder with the base64-encoded PEM of the
# certctl-server's self-signed bootstrap root, then `kubectl apply`.
#
# Generate the caBundle via:
# cat deploy/test/certs/ca.crt | base64 -w0
# (See certctl/docs/acme-server.md "TLS trust bootstrap" section for the
# end-to-end walkthrough — this is the single biggest first-time-deploy
# footgun on cert-manager, captured as audit fix #9.)
apiVersion: cert-manager.io/v1
kind: ClusterIssuer
metadata:
name: certctl-test-trust
spec:
acme:
email: test@example.com
# Replace 'certctl-test' with your release name + adjust the
# profile path segment. Default profile path:
# https://<service>.<namespace>.svc.cluster.local:8443/acme/profile/<profile-id>/directory
server: https://certctl-test.default.svc.cluster.local:8443/acme/profile/prof-test/directory
# caBundle: Audit fix #9. cert-manager validates the ACME server's
# TLS chain before submitting any account/order/finalize. With a
# self-signed bootstrap root, the ClusterIssuer MUST carry the root
# explicitly via this field.
caBundle: |
LS0tLS1CRUdJTiBDRVJUSUZJQ0FURS0tLS0tCi4uLgotLS0tLUVORCBDRVJUSUZJQ0FURS0tLS0tCg==
privateKeySecretRef:
name: certctl-test-trust-account-key
solvers:
# In trust_authenticated mode the solver is unused at the
# validation step but cert-manager still requires at least one
# solver in the spec. http01-via-ingress-nginx is the cheapest
# placeholder shape that round-trips correctly through cert-
# manager's validation webhooks.
- http01:
ingress:
class: nginx
```
This block is byte-equal to
[`deploy/test/acme-integration/clusterissuer-trust-authenticated.yaml`](../deploy/test/acme-integration/clusterissuer-trust-authenticated.yaml).
Replace the `caBundle` placeholder with the base64 string from Step 3.
The full reference YAML lives at
[`deploy/test/acme-integration/clusterissuer-trust-authenticated.yaml`](../deploy/test/acme-integration/clusterissuer-trust-authenticated.yaml).
```
kubectl apply -f deploy/test/acme-integration/clusterissuer-trust-authenticated.yaml
kubectl wait --for=condition=Ready --timeout=2m clusterissuer/certctl-test-trust
```
The solver block is a placeholder under `trust_authenticated` mode —
cert-manager 1.15 still requires at least one solver in the spec, but
certctl auto-resolves authzs without a solver round-trip. The
http01-ingress-nginx shape validates against cert-manager's webhook
without needing an actual ingress controller deployed.
For `challenge` mode profiles, swap to
[`deploy/test/acme-integration/clusterissuer-challenge.yaml`](../deploy/test/acme-integration/clusterissuer-challenge.yaml)
— same shape, but the solver is now load-bearing and you need
ingress-nginx (or your chosen ingress class) actually deployed for
HTTP-01 to work.
## Step 5 — Apply the Certificate
```yaml
# Phase 5 — Certificate resource the integration test applies and
# waits for. The certctl-test-trust ClusterIssuer (trust_authenticated
# mode) issues the cert without any solver round-trip; the resulting
# Secret 'test-com-tls' is asserted to carry tls.crt + tls.key.
apiVersion: cert-manager.io/v1
kind: Certificate
metadata:
name: test-com
namespace: default
spec:
secretName: test-com-tls
commonName: test.example.com
dnsNames:
- test.example.com
- www.test.example.com
issuerRef:
name: certctl-test-trust
kind: ClusterIssuer
duration: 720h # 30d
renewBefore: 240h # 10d
```
This block is byte-equal to
[`deploy/test/acme-integration/certificate-test.yaml`](../deploy/test/acme-integration/certificate-test.yaml).
```
kubectl apply -f deploy/test/acme-integration/certificate-test.yaml
kubectl wait --for=condition=Ready --timeout=3m certificate/test-com
```
cert-manager creates an `Order`, the ACME flow runs against certctl,
and the resulting Secret is populated.
## Step 6 — Verify
```
kubectl get certificate test-com -o wide
# NAME READY SECRET ISSUER STATUS AGE
# test-com True test-com-tls certctl-test-trust Certificate is up to date and has not expired 42s
kubectl get secret test-com-tls -o yaml | yq '.data."tls.crt"' | base64 -d | openssl x509 -noout -subject -issuer -dates
# subject= CN=test.example.com
# issuer= CN=certctl test internal CA
# notBefore=... notAfter=...
```
Both the cert-manager `Certificate` resource and the underlying Secret
are populated. The actor on the certctl side is `acme:<account-id>`,
which you can correlate via the `audit_events` table:
```
psql -c "SELECT created_at, action, resource_type, resource_id
FROM audit_events
WHERE actor LIKE 'acme:%'
ORDER BY created_at DESC LIMIT 10;"
```
## Common failure modes
These are operator-side; full troubleshooting reference is in
[`docs/acme-server.md` § Troubleshooting](./acme-server.md#troubleshooting).
- `400 Bad Request: badNonce` → clock skew between certctl-server and
cert-manager, or a multi-replica certctl fleet without sticky
sessions.
- `x509: certificate signed by unknown authority` → missing or stale
`caBundle`. Re-run Step 3, paste the fresh value.
- `connection refused` from the HTTP-01 validator → ingress controller
not deployed, OR your network blocks port 80 inbound to the solver
Ingress.
- `Ready=False` with `rejectedIdentifier` → CSR has a SAN your profile
policy doesn't permit. Decode the `subproblems` array of the RFC
7807 problem doc.
## Cleanup
```
kubectl delete -f deploy/test/acme-integration/certificate-test.yaml
kubectl delete -f deploy/test/acme-integration/clusterissuer-trust-authenticated.yaml
helm uninstall certctl-test
# Optional: delete the certctl profile via API.
```
## See also
- [`docs/acme-server.md`](./acme-server.md) — canonical reference.
- [`docs/acme-server-threat-model.md`](./acme-server-threat-model.md) —
security posture.
- [`docs/acme-caddy-walkthrough.md`](./acme-caddy-walkthrough.md) —
Caddy-side recipe.
- [`docs/acme-traefik-walkthrough.md`](./acme-traefik-walkthrough.md) —
Traefik-side recipe.
- [`deploy/test/acme-integration/`](../deploy/test/acme-integration/) —
Phase 5 integration test (the same recipe, automated).
+205
View File
@@ -0,0 +1,205 @@
# Traefik Integration Walkthrough
> **Use this walkthrough when** you're already running Traefik 3.0+
> (Kubernetes or VM) and want it to ACME-issue from certctl (your
> internal CA, your private PKI, or a local sub-CA chained under an
> enterprise root) instead of Let's Encrypt. The Traefik static config
> changes are minimal; the load-bearing piece is `serversTransport.rootCAs`
> so Traefik trusts certctl's bootstrap CA on every outbound ACME call.
End-to-end recipe for issuing certs from a certctl-server deployment
through Traefik 3.0+. Target audience: operator running Traefik (in
Kubernetes or on a VM) who wants to use certctl as their ACME source
of truth instead of Let's Encrypt.
## Prereqs
- A reachable certctl-server with `CERTCTL_ACME_SERVER_ENABLED=true`
and at least one profile whose `acme_auth_mode` is set. Profile
setup is identical to the cert-manager walkthrough — see
[`docs/acme-cert-manager-walkthrough.md`](./acme-cert-manager-walkthrough.md)
Step 2.
- Traefik 3.0+ (the v2 API surface for ACME is also supported but the
`serversTransport.rootCAs` reference below is v3-shaped).
- The certctl bootstrap CA, in PEM form, captured the same way as the
cert-manager walkthrough Step 3.
## Step 1 — Configure Traefik static config
Traefik's ACME issuer is a `certificatesResolver` in the static config
(file or CLI flags or env vars). The relevant fields:
```yaml
# /etc/traefik/traefik.yml (or wherever your static config lives)
certificatesResolvers:
certctl:
acme:
caServer: https://certctl.example.com:8443/acme/profile/prof-test/directory
email: ops@example.com
storage: /etc/traefik/acme-certctl.json
httpChallenge:
entryPoint: web
# OR for trust_authenticated mode profiles:
# tlsChallenge: {}
# certctl uses a self-signed bootstrap cert; Traefik needs the CA
# explicitly via serversTransport.rootCAs to call the directory URL.
serversTransports:
default:
rootCAs:
- /etc/traefik/certctl-bootstrap.crt
# Apply the serversTransport globally so every outbound HTTPS call —
# including ACME directory + finalize — trusts the certctl CA.
api:
insecure: false
entryPoints:
web:
address: ":80"
websecure:
address: ":443"
```
Notes:
- `caServer` must point at the directory URL (ending in `/directory`).
- `httpChallenge.entryPoint: web` requires Traefik's `web` entryPoint
(port 80) to be reachable from certctl-server's HTTP-01 validator.
For `trust_authenticated` mode profiles, this is a no-op formality —
certctl auto-resolves authzs, so the solver round-trip never happens.
- `tlsChallenge: {}` is the alternative that uses TLS-ALPN-01 (RFC 8737)
via Traefik's `websecure` (port 443) entryPoint. Either works under
`challenge` mode; only the default-of-`tlsChallenge` is recommended
for `trust_authenticated` mode.
## Step 2 — Trust the certctl bootstrap CA
Two options:
### Option A — `serversTransport.rootCAs` (preferred)
```
sudo cp deploy/test/certs/ca.crt /etc/traefik/certctl-bootstrap.crt
sudo systemctl reload traefik
```
`serversTransports.default.rootCAs` (shown in Step 1 above) tells
Traefik's outbound HTTPS client to trust the supplied PEM in addition
to the system trust store. This is the right pattern for containerized
Traefik where you don't want to install OS-level trust roots.
### Option B — OS trust store
For Traefik running directly on a VM, `update-ca-certificates`-style
installation works the same way as the Caddy walkthrough Option A.
The `serversTransport.rootCAs` field is unnecessary in that case.
## Step 3 — Reference the resolver from a router
Per-router (dynamic config):
```yaml
# /etc/traefik/dynamic/example-com.yml
http:
routers:
example-com:
rule: "Host(`example.com`)"
entryPoints: [websecure]
tls:
certResolver: certctl
service: example-com-backend
services:
example-com-backend:
loadBalancer:
servers:
- url: "http://localhost:8080"
```
Or, in Kubernetes via `IngressRoute` (Traefik CRD):
```yaml
apiVersion: traefik.io/v1alpha1
kind: IngressRoute
metadata:
name: example-com
spec:
entryPoints: [websecure]
routes:
- match: Host(`example.com`)
kind: Rule
services:
- name: example-com-backend
port: 8080
tls:
certResolver: certctl
```
## Step 4 — Reload Traefik
```
sudo systemctl reload traefik
# OR kubectl rollout restart deployment/traefik (if you changed the static config via ConfigMap).
```
On the first request to `example.com`, Traefik hits certctl's directory
URL, registers an account, submits a new-order, and finalizes. The cert
is persisted to `/etc/traefik/acme-certctl.json` (or its in-cluster
PVC equivalent).
## Step 5 — Verify
```
curl -kvI https://example.com 2>&1 | grep -E 'subject|issuer'
# subject: CN=example.com
# issuer: CN=certctl test internal CA
```
The cert is signed by certctl's bound issuer (per the `prof-test`
profile's `issuer_id`).
On the certctl side, the audit log captures the issuance:
```
psql -c "SELECT actor, action, resource_id FROM audit_events
WHERE actor LIKE 'acme:%' ORDER BY created_at DESC LIMIT 5;"
```
## Common failure modes
- **Traefik logs `unable to obtain ACME certificate ... x509: certificate
signed by unknown authority`** → `serversTransport.rootCAs` is not
pointing at the certctl bootstrap CA, OR the file was rotated and
Traefik hasn't reloaded. Verify with
`curl --cacert /etc/traefik/certctl-bootstrap.crt
https://certctl.example.com:8443/acme/profile/prof-test/directory`.
- **Traefik logs `urn:ietf:params:acme:error:rateLimited`** → tune
`CERTCTL_ACME_SERVER_RATE_LIMIT_ORDERS_PER_HOUR` on the certctl
side, OR reduce Traefik's parallel-cert-acquisition concurrency.
- **`acme: error: 400 :: POST :: ... :: badNonce`** → clock skew or
multi-replica certctl without sticky sessions; same fix as the
cert-manager walkthrough.
- **Storage file `acme-certctl.json` shows persistent failures** —
Traefik retains failed-acquisition state. After fixing the
underlying cause, delete the storage file and reload:
`rm /etc/traefik/acme-certctl.json && systemctl reload traefik`.
## Cleanup
```
# Remove the certResolver from any router / IngressRoute consuming it.
sudo systemctl reload traefik
# Delete the persisted ACME storage:
sudo rm /etc/traefik/acme-certctl.json
# Or in K8s: drop the resolver from the static-config ConfigMap.
```
## See also
- [`docs/acme-server.md`](./acme-server.md) — canonical reference.
- [`docs/acme-cert-manager-walkthrough.md`](./acme-cert-manager-walkthrough.md) —
cert-manager equivalent.
- [Traefik upstream ACME docs](https://doc.traefik.io/traefik/https/acme/#caserver) —
verify behavior pinned here against Traefik 3.0+ semantics.