Follow-up tocfc234e(U-1 docker-compose fix) — closes the remaining adjacent code paths that share the postgres-first-boot-password-binding root cause but were scoped out of the original commit. The runtime diagnostic in internal/repository/postgres/db.go::wrapPingError (landed ina911970) already covers every NewDB call site, so Helm operators and example users hit the SQLSTATE 28P01 guidance for free at startup. What was missing: deployment-shape-specific remediation guidance (kubectl vs docker-compose), the hardcoded password in the *root* .env.example, and shared ops notes for the 5 examples/ compose files. This commit closes all three. Files changed: - .env.example (root) — line 16 had `postgres://certctl:certctl@...` with the password hardcoded literally instead of interpolating POSTGRES_PASSWORD. Edit if a user copied this file as their .env (binary-direct deployment, not docker-compose) and rotated POSTGRES_PASSWORD on line 10, the URL on line 16 still carried 'certctl' — silent two-line drift. Replaced 'certctl' with the same default that line 10 carries ('change-me-in-production') and added an explanatory comment block describing the docker-compose override semantics, when this URL matters (binary-direct), and the cross-reference to the U-1 wrapPingError diagnostic. Also fixed an adjacent bug: line 31 CERTCTL_SERVER_URL was `http://localhost:8443`, which agents reject at startup since v2.2 (HTTPS-everywhere milestone made the control plane HTTPS-only with TLS 1.3 pinned). Updated to https:// with a comment pointing operators at the bootstrap CA bundle. - deploy/helm/certctl/values.yaml — postgresql.auth.password field had a one-line 'REQUIRED' comment. Expanded into a full WARNING block (~25 lines) explaining the PVC retention semantics, the failure symptom, and both kubectl-flavored remediation paths: non-destructive (`kubectl exec ... ALTER ROLE`) preferred for environments with data, and destructive (`helm uninstall + kubectl delete pvc`) for dev/demo. Cross-references the wrapPingError runtime diagnostic. - deploy/helm/certctl/README.md (new, ~115 lines) — chart-level operational guide. Covers quick install, both remediation paths with concrete kubectl commands, why-we-don't-fix-this-in-the-chart explanation, cross-references to the docker-compose docs, server API key rotation (the easy case — comma-separated key list), TLS provisioning shapes, embedded-vs-external postgres, and uninstall semantics with the PVC retention gotcha called out. - examples/README.md (new, ~55 lines) — shared operational notes for the 5 example deployments. Covers the postgres password rotation trap with example-flavored remediation paths (`docker compose -f examples/<x>/...`), the TLS warning, and teardown semantics. Replaces what would otherwise be 5x duplication across per-example READMEs. - examples/{acme-nginx,acme-wildcard-dns01,multi-issuer,private-ca-traefik, step-ca-haproxy}/*.md — one-line cross-reference at the top of each example's primary doc, pointing at examples/README.md for the shared ops notes. Avoids 5x duplication of the same warning text while still surfacing the link in every operator's first-touch surface. Verification: - go build ./... — clean - go vet ./... — clean - go test -short ./internal/repository/postgres/ — 4/4 wrapPingError tests still passing (no production-code touch in this commit) - helm lint deploy/helm/certctl/ — clean (1 INFO about chart icon, pre-existing) - helm template smoke test — renders without error - python3 yaml.safe_load on values.yaml — parses Refs: coverage-gap-audit-2026-04-24-v5/unified-audit.md §2 P1 cluster, cat-u-quickstart_postgres_password_volume_trap Closes the three deliberate scope-outs fromcfc234e(Helm, root .env.example, examples/) end-to-end. Adjacent bugs caught while in scope: - root .env.example:16 hardcoded password not matching line 10 - root .env.example:31 http:// URL incompatible with HTTPS-only v2.2
10 KiB
step-ca + HAProxy Example
This example demonstrates certctl managing certificates issued by Smallstep step-ca and deploying them to HAProxy.
Operational notes shared by every example (postgres password rotation trap, TLS provisioning, teardown semantics) live in
../README.md. Read it first if you plan to changeDB_PASSWORDafter the initialdocker compose up— the postgres volume binds the password on first boot only.
Scenario
You're a Smallstep user running step-ca as your internal PKI. You have HAProxy load balancers that need certificates. This setup:
- step-ca issues certificates (via JWK provisioner, no challenge solving)
- certctl manages the certificate lifecycle (renewal policies, deployment, audit)
- HAProxy serves HTTPS with certificates managed by certctl
This is the natural choice if you're already invested in step-ca and want to consolidate certificate lifecycle management without learning Let's Encrypt, DNS-01 challenges, or external integrations.
What's Included
| Service | Image | Purpose |
|---|---|---|
| step-ca | smallstep/step-ca:latest |
Private internal CA |
| certctl-server | ghcr.io/shankar0123/certctl-server:latest |
Certificate management control plane |
| certctl-agent | ghcr.io/shankar0123/certctl-agent:latest |
Agent running on HAProxy server |
| haproxy | haproxy:2.9-alpine |
Reverse proxy / load balancer |
| postgres | postgres:16-alpine |
certctl audit trail + config storage |
Quick Start
Prerequisites
- Docker and Docker Compose
- Curl (to interact with APIs)
1. Start Everything
docker compose up -d
This will:
- Initialize step-ca with a self-signed root CA
- Create a JWK provisioner named
certctl(pre-configured credentials) - Start certctl-server (connected to step-ca)
- Start the certctl-agent (ready to deploy certs to HAProxy)
- Start HAProxy with a placeholder config
Monitor logs:
docker compose logs -f certctl-server
TLS Security
certctl is HTTPS-only as of v2.2. The demo compose stack provisions a self-signed certificate. When accessing https://localhost:8443, you can either:
- Use
curl --cacert ./deploy/test/certs/ca.crt ...to pin the CA certificate - Use
curl -k ...for quick smoke tests (never in production) - Import the CA at
./deploy/test/certs/ca.crtinto your OS trust store for browser visits
Wait for all services to reach healthy state:
docker compose ps
Expected output:
NAME STATUS
certctl-postgres-... healthy
certctl-server-... healthy
step-ca-... healthy
certctl-agent-... running
certctl-haproxy-... healthy
2. Access certctl Dashboard
Open your browser to:
https://localhost:8443
You should see an empty dashboard. This is expected — no certificates issued yet.
3. Create a Certificate Profile
This defines what certificates certctl can issue (key algorithm, max TTL, allowed names).
curl -X POST https://localhost:8443/api/v1/profiles \
-H 'Content-Type: application/json' \
-d '{
"name": "internal-web",
"key_type": "rsa-2048",
"max_ttl_days": 90,
"description": "Internal web services"
}'
4. Create an HAProxy Deployment Target
This tells certctl where to deploy certificates on the HAProxy server.
curl -X POST https://localhost:8443/api/v1/targets \
-H 'Content-Type: application/json' \
-d '{
"name": "haproxy-01",
"type": "haproxy",
"enabled": true,
"config": {
"pem_path": "/etc/haproxy/ssl/cert.pem",
"reload_command": "systemctl reload haproxy",
"validate_command": "haproxy -c -f /etc/haproxy/haproxy.cfg"
}
}'
Note: In the Docker Compose environment, reload command can be kill -HUP $(pidof haproxy) instead of systemctl reload haproxy.
5. Create a Renewal Policy
This ties a certificate profile to a deployment target and sets renewal thresholds.
curl -X POST https://localhost:8443/api/v1/renewal-policies \
-H 'Content-Type: application/json' \
-d '{
"name": "haproxy-internal-web",
"profile_id": "<profile_id_from_step_3>",
"issuer_id": "iss-stepca",
"enabled": true,
"renewal_days_before_expiry": 30,
"alert_thresholds_days": [30, 14, 7, 0]
}'
Get the issuer ID:
curl https://localhost:8443/api/v1/issuers | jq '.'
You should see iss-stepca in the list.
6. Issue a Certificate
Request a certificate via the API. The server will sign it via step-ca.
curl -X POST https://localhost:8443/api/v1/certificates \
-H 'Content-Type: application/json' \
-d '{
"common_name": "api.internal.example.com",
"sans": ["api.internal.example.com", "api.staging.example.com"],
"issuer_id": "iss-stepca",
"profile_id": "<profile_id_from_step_3>"
}'
7. Deploy to HAProxy
Get the certificate ID and trigger deployment:
curl -X POST https://localhost:8443/api/v1/certificates/<cert_id>/deploy \
-H 'Content-Type: application/json' \
-d '{
"target_id": "<target_id_from_step_4>"
}'
The agent will:
- Fetch the deployment job
- Generate a combined PEM (cert + chain + key) locally
- Write it to
/etc/haproxy/ssl/cert.pemon HAProxy - Reload HAProxy
- Report status back to certctl
8. Verify in Dashboard
Refresh https://localhost:8443 and you should see:
- 1 certificate (status: Active, expiry in 90 days)
- 1 deployment job (status: Completed)
- 1 agent (heartbeat: recent)
Configuration Details
step-ca Integration
step-ca is configured with:
- Root CA Name:
certctl-demo-ca - Provisioner:
certctl(JWK type) - Default Password:
certctl-provisioner-demo(override withSTEP_CA_PROVISIONER_PASSWORD)
To inspect step-ca:
docker compose exec step-ca step ca provisioner list
docker compose exec step-ca step ca health --insecure
HAProxy Combined PEM Format
HAProxy requires a single file with certificate, chain, and key concatenated:
-----BEGIN CERTIFICATE-----
[leaf certificate]
-----END CERTIFICATE-----
-----BEGIN CERTIFICATE-----
[intermediate CA]
-----END CERTIFICATE-----
-----BEGIN RSA PRIVATE KEY-----
[private key]
-----END RSA PRIVATE KEY-----
The agent automatically constructs this file from the issued certificate and step-ca-provided chain.
Security: The combined PEM is written with 0600 permissions (owner-readable only) because it contains the private key.
Environment Variables
Customize behavior with:
| Variable | Default | Purpose |
|---|---|---|
DB_PASSWORD |
certctl-dev-password |
PostgreSQL password |
STEP_CA_PASSWORD |
stepca-demo-password |
step-ca root key password |
STEP_CA_PROVISIONER_PASSWORD |
certctl-provisioner-demo |
certctl JWK provisioner password |
AGENT_API_KEY |
agent-demo-key |
Agent authentication token |
SERVER_PORT |
8443 |
certctl server external port |
Example:
STEP_CA_PASSWORD=myca-password AGENT_API_KEY=secret-key docker compose up -d
Integrating with an Existing step-ca Instance
If you already run step-ca elsewhere (not in this Compose file):
-
Extract the root certificate from your step-ca:
step ca root /tmp/step-ca-root.crt --ca-url https://ca.internal:9000 --insecure -
Create or retrieve the certctl JWK provisioner key:
step ca provisioner list --ca-url https://ca.internal:9000 --insecure step ca provisioner describe certctl --ca-url https://ca.internal:9000 --insecure -
Update docker-compose.yml:
certctl-server: environment: CERTCTL_STEPCA_URL: https://ca.internal:9000 CERTCTL_STEPCA_ROOT_CERT_PATH: /etc/certctl/step-ca-root.crt CERTCTL_STEPCA_PROVISIONER_NAME: certctl CERTCTL_STEPCA_PROVISIONER_KEY_PATH: /etc/certctl/step-ca-provisioner.json CERTCTL_STEPCA_PROVISIONER_PASSWORD: <your-password> -
Mount the cert and key:
volumes: - /path/to/step-ca-root.crt:/etc/certctl/step-ca-root.crt:ro - /path/to/provisioner.json:/etc/certctl/step-ca-provisioner.json:ro
Cleanup
docker compose down -v
This removes all containers and volumes (step-ca config, certificates, database).
Next Steps
Production Deployment
- Replace image tags (
latest→ specific version) - Use real TLS certificates for step-ca (self-signed is fine internally, but use proper roots for verification)
- Configure persistent storage for step-ca keys (HSM or encrypted filesystem)
- Set
CERTCTL_AUTH_TYPE: api-keyand rotate API keys regularly - Enable audit trail export for compliance
- Configure renewal alerts (Slack, email, PagerDuty)
- Run agents on separate machines (not in Compose)
Advanced Features
- Multiple HAProxy instances: Create additional targets and agents
- Policy-based renewal: Set different renewal windows per environment (staging vs. production)
- Approval workflows: Require manual approval before deploying to production
- Discovery: Scan existing HAProxy certs and bring them under management
- Network scanning: Discover TLS endpoints in your network and inventory them
Troubleshooting
step-ca fails to initialize
Check logs:
docker compose logs step-ca
Common issues:
- Permissions on
/home/step/step-cavolume - Port 9000 already in use
Agent can't reach server
Verify network:
docker compose exec certctl-agent curl http://certctl-server:8443/health
HAProxy config validation fails
Check HAProxy config syntax:
docker compose exec haproxy haproxy -c -f /etc/haproxy/haproxy.cfg
Deployment job stays in "Running" state
Check agent logs:
docker compose logs certctl-agent
Likely causes:
- Agent can't write to
/etc/haproxy/ssl/cert.pem(permissions) - Reload command is misconfigured
- HAProxy container is not accessible
Documentation
Support
For issues or questions:
- Check the troubleshooting guide
- Review service logs:
docker compose logs <service> - Open an issue on GitHub