blumeops/docs/how-to/zot/harden-zot-registry.md
Erich Blume c86b5d7772
All checks were successful
Build Container / detect (push) Successful in 3s
Build Container / build-dagger (navidrome) (push) Successful in 22m26s
Native Dagger container builds + Navidrome v0.61.1 (#330)
## Summary
- Move Dagger module from `.dagger/` to repo root (`src/blumeops/`), rename `blumeops-ci` → `blumeops`
- Replace opaque `docker_build()` with native Dagger pipelines that surface full build errors per step
- Migrate navidrome as the first container (`containers/navidrome/container.py`)
- Upgrade navidrome from v0.60.3 to v0.61.1 (major artwork overhaul, SQLite FTS5 search, server-managed transcoding)
- Add `dagger call container-version` for CI version extraction without Dockerfile parsing
- All mise tasks (`container-list`, `container-version-check`, `container-build-and-release`) updated for hybrid mode
- Legacy `docker_build()` fallback preserved for all other containers

## Motivation
When navidrome v0.61.0 added a new Go build tag (`sqlite_fts5`), `docker_build()` showed only "exit code: 1". We had to run `docker build --progress=plain` manually to find `undefined: buildtags.SQLITE_FTS5`. Native Dagger pipelines show the full error inline.

## Container build dispatch needed
After merge, dispatch container build for navidrome:
```
mise run container-build-and-release navidrome --ref 470b4bd
```

## Deploy steps
1. Wait for container build to complete
2. Back up navidrome-data PVC (non-reversible DB migrations)
3. `argocd app set navidrome --revision main && argocd app sync navidrome`
4. Verify at https://dj.ops.eblu.me

## Future
Remaining containers migrate incrementally in follow-up PRs using the same pattern.

Reviewed-on: #330
2026-04-11 17:11:56 -07:00

55 lines
2.2 KiB
Markdown

---
title: Harden Zot Registry
modified: 2026-04-11
tags:
- how-to
- zot
- registry
- security
---
# Harden Zot Registry
OIDC + API key authentication on zot with anonymous pull preserved, and tag immutability enforced server-side via accessControl. Completed as a C2 Mikado goal across PRs #236 and #237.
## What Was Done
Updated `ansible/roles/zot/templates/config.json.j2` with:
1. **`http.auth.openid`** — OIDC provider pointing to Authentik (`authentik.ops.eblu.me`)
2. **`http.auth.apikey: true`** — API key generation for CI service accounts
3. **`http.accessControl`** — three-tier policy:
- `anonymousPolicy: ["read"]` — anyone can pull
- `artifact-workloads` group: `["read", "create"]` — CI can push new tags but cannot overwrite or delete (immutable tags)
- `admins` group: `["read", "create", "update", "delete"]` — break-glass
4. **`http.externalUrl`** — `https://registry.ops.eblu.me` for OIDC callback redirects
5. **`accessControl.metrics.users: [""]`** — allows anonymous Prometheus/Alloy scraping
## Key Files
| File | Purpose |
|------|---------|
| `ansible/roles/zot/templates/config.json.j2` | Zot config with auth + access control |
| `ansible/roles/zot/defaults/main.yml` | OIDC issuer and external URL variables |
| `ansible/roles/zot/templates/oidc-credentials.json.j2` | OIDC client credentials |
| `src/blumeops/main.py` | `publish()` with registry auth |
| `.forgejo/workflows/build-container.yaml` | Dagger push with API key |
| `.forgejo/workflows/build-container-nix.yaml` | Skopeo push with API key |
## Verified
- [x] Anonymous pull works (pull-through cache on gilbert)
- [x] Unauthenticated push fails (401)
- [x] OIDC browser login works (redirect to Authentik and back)
- [x] API key push works (zot-ci API key)
- [x] CI push succeeds (Dagger and Nix/skopeo paths)
- [x] Pull-through caching still works
- [x] Metrics endpoint accessible without auth
- [x] `mise run services-check` passes
## Related
- [[register-zot-oidc-client]] — OIDC client registration in Authentik
- [[wire-ci-registry-auth]] — CI push path wiring
- [[enforce-tag-immutability]] — Server-side via accessControl
- [[adopt-commit-based-container-tags]] — Commit-SHA-based image tags