The heph CLI requests scope "openid offline_access", but the Authentik
heph OAuth2 provider only mapped openid/email/profile. Without the
offline_access mapping the issued refresh token is bound to the login
session rather than the 30-day refresh-token window; once the session
lapses, hephd's refresh_token grant returns 400 Bad Request and spoke
sync silently degrades (heph sync --status -> auth_failure: true).
Add the built-in offline_access scope mapping to the provider's
property_mappings and document the requirement in the service reference.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Adds the heph-pwa redirect URIs to the Authentik `heph` OAuth2 provider so the new browser **Login with Authentik** flow (Authorization Code + PKCE, hephaestus PR #9) can redirect back and exchange the code:
- `https://heph.ops.eblu.me/` (the PWA origin)
- `http://localhost:8787/` (local dev: `hephd --web-root`)
Authentik also keys token-endpoint CORS off these origins, so they're required for the browser token exchange. Additive (the provider was `redirect_uris: []`); harmless until the PWA feature deploys.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #370
Makes indri the canonical **heph** hub for the hub-and-spoke task/context system, deployed as a self-updating LaunchAgent managed by Ansible. Other devices (gilbert) attach as offline-capable spokes.
## What's here
- **`ansible/roles/heph`** (tag `heph`) — bootstrap `cargo install hephd` (only if absent; `--self-update` keeps it current after), version-pinned `heph-pwa` checkout served via `--web-root`, launchagent `mcquack.eblume.heph`:
```
hephd --mode server --http-addr 0.0.0.0:8787 --db … --web-root …
--oidc-issuer …/o/heph/ --oidc-audience heph
--self-update --self-update-interval-secs 600
```
`~/.cargo/bin` is on the agent `PATH` so self-update's `cargo install` works.
- **Caddy** — `heph.ops.eblu.me → localhost:8787` (TLS for the PWA secure context).
- **Authentik** — new `heph` **public device-code** OIDC app + `default-device-code-flow` bound to the default brand's `flow_device_code` (verified live: brand `authentik-default`, field currently unset → additive).
- **Docs** — `services/hephaestus.md` (Path-A seeding runbook + spoke caveat), `indri.md`, changelog fragment.
## Three features requested
- **Autoupdate** — 10-min interval (`--self-update-interval-secs 600`).
- **PWA** — `--web-root` (confirmed shipped in v1.2.0).
- **Spoke** — gilbert reconfig documented (post-merge step).
## Deploy plan (not done yet — awaiting review)
1. Seed from gilbert (Path A): `heph daemon stop` → copy `heph.db` → `DELETE FROM meta WHERE key='origin'`.
2. Sync Authentik `apps`/blueprint; verify blueprint status via API (not just logs).
3. `provision-indri --tags heph,caddy` from this branch.
4. Point gilbert at the hub + `heph auth login`.
## Known follow-ups (heph-side, tracked in the Hephaestus project)
- `heph daemon` can't bake hub/spoke config or pass `--self-update-interval-secs` → worked around by the ansible plist.
- Path-A seeding lacks a clean `hephd --owner-id`/seed command → manual `meta.origin` reset for now.
- Self-update moves hephd ahead of the ansible-pinned PWA shell over time (drift; tolerated by the SW cache, revisit on next release).
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #369
indri -> v2.2.0-13895bb (arm64), ringtail -> v2.2.0-13895bb-nix (amd64).
Both deployed images now trace to main commit 13895bb instead of earlier
branch builds.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Follow-up to #367. That PR localized external-secrets but the Dagger build (on indri's Apple Silicon runner) only produces an **arm64** image — and external-secrets also runs on **ringtail (amd64)** via the same shared manifest. This completes the localization so both clusters run the local binary on their native arch.
## Approach (matches the kube-state-metrics dual-build pattern)
- **`containers/external-secrets/default.nix`** (new) — builds the **amd64** image on ringtail's nix-container-builder. `buildGoModule` with Go 1.26 (v2.2.0 requires ≥1.26.1; nixpkgs default is 1.25.x) and `-tags all_providers`, faithful to upstream. Same v2.2.0 source from the forge mirror.
- **`argocd/manifests/external-secrets-ringtail/`** (new) — thin kustomize overlay that reuses the shared indri manifest as a base and overrides **only** the image to the `-nix` (amd64) tag. No manifest duplication.
- **`argocd/apps/external-secrets-ringtail.yaml`** — repointed at the new overlay.
Result: indri → `v2.2.0-…` (arm64, Dagger), ringtail → `v2.2.0-…-nix` (amd64, nix).
## Build
Run #581 built both arches at the branch commit. Verified the nix image is `linux/amd64`, entrypoint = the binary, user 65534.
## Deployed from branch & verified on ringtail (k3s, amd64)
- All 3 pods rolled to the nix amd64 image, `1/1 Running` (no exec-format error → arch correct)
- Controller logs clean
- **Live secret fetch proven:** force-synced `homepage/homepage-grafana` → `refreshTime` advanced, `Ready=True`
- **All 20** ringtail ExternalSecrets remain `SecretSynced=True`
## Post-merge
The `external-secrets-ringtail` app is temporarily pointed at this branch + overlay path (apps app left on `main`, manual-sync, untouched). After merge:
```
argocd app sync apps # picks up the new Application path on main
argocd app set external-secrets-ringtail --revision main && argocd app sync external-secrets-ringtail
```
I'll also rebuild off `main` so both clusters land on stable main-sha tags (as done for indri in #367).
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #368
Repoint to the main-branch-built image so the deployed tag traces to a main
commit rather than the merged feature branch. Same v2.2.0 source, stable
provenance.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Knocks out the weekly "pick one non-local container and make it local" task by moving **external-secrets** off `ghcr.io` onto a locally-built image, under our own supply-chain control. Doubles as its overdue service review.
## What changed
- **`containers/external-secrets/container.py`** (new) — native Dagger build (the Dockerfile→container.py migration pattern). Clones the forge mirror at `v2.2.0` and builds the single `all_providers` static Go binary, faithful to upstream's `make build` (CGO off, no version ldflags upstream). ENTRYPOINT is `/bin/external-secrets` so the controller/webhook/cert-controller Deployments select their role via `args:` exactly as before.
- **`argocd/manifests/external-secrets/kustomization.yaml`** — image swapped to `registry.ops.eblu.me/blumeops/external-secrets:v2.2.0-2985007`. **Like-for-like (v2.2.0)**, not an upgrade.
- **`service-versions.yaml`** — marked reviewed (2026-06-04), noted the local build.
## Build
Built on the indri forge runner (run #579, ~4 min) → pushed to Zot. Image config verified: `Entrypoint=/bin/external-secrets`, `User=65534`, version label `v2.2.0`.
## Deployed from branch & verified
- All 3 pods (controller / webhook / cert-controller) rolled to the local image, `1/1 Running`
- Controller + webhook logs clean (no errors; webhook serving TLS)
- **End-to-end secret fetch proven:** force-synced `monitoring/grafana-admin` → `refreshTime` advanced to now, `Ready=True`
- All 10 ExternalSecrets cluster-wide remain `SecretSynced=True` — no collateral damage
- App `Healthy`
## Post-merge
`external-secrets` currently points at this branch (so `apps` reads OutOfSync — expected). After merge:
```
argocd app set external-secrets --revision main && argocd app sync external-secrets
```
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #367
Knocks out the two daily recurring review tasks (doc review + service review) in one PR.
## Doc review (4 never-reviewed reference cards, `last-reviewed: 2026-06-04`)
- **cluster.md** — Kubernetes version v1.34.0 → **v1.35.0**; refreshed the stale ringtail workload list and noted the in-progress minikube→k3s migration (points to `[[ringtail]]` as the canonical list).
- **ntfy.md / tempo.md / alloy.md** — corrected image references: these are now **locally-built `registry.ops.eblu.me/blumeops/*` nix containers** (ntfy v2.19.2, tempo v2.10.3, alloy-k8s v1.16.0), not upstream Docker Hub. Fly.io alloy binary bumped to v1.16.1.
## Service review
- **nvidia-device-plugin** (ringtail GPU): v0.19.0 → **v0.19.2**. Upstream patch releases — CDI/Tegra fixes + dependency bumps, no breaking changes for our manifest-based CDI + RuntimeClass setup (the service-account change in the notes is helm-only).
## Not in this PR (need container rebuilds, deferred)
The other stale services are locally-built nix images, so upgrading them is a forge-runner rebuild rather than a clean tag bump — left untouched (not date-bumped, so they resurface): **prometheus** (v3.10.0→v3.12.0), **loki** (3.6.7→3.7.2), **kube-state-metrics**, **homepage**. Happy to do these as a follow-up rebuild PR.
## Deploy / verify
Not yet deployed — `nvidia-device-plugin` still points at `main`. After review:
```
argocd app set nvidia-device-plugin --revision reviews-jun4 && argocd app sync nvidia-device-plugin
# after merge:
argocd app set nvidia-device-plugin --revision main && argocd app sync nvidia-device-plugin
```
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #366
The public forge.eblu.me now black-holes /mirrors/ at the Fly edge
(AI-scraper mitigation), so the in-cluster ArgoCD repo-server got a 403
fetching the upstream operator manifest — leaving tailscale-operator and
tailscale-operator-ringtail in Unknown sync. Use forge.ops.eblu.me.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The Dagger build_docs pipeline cloned Quartz from the default branch
unpinned. Quartz v5.0.0 restructured its config layout (.quartz/plugins,
../quartz imports), breaking the docs build against our existing
quartz.config.ts / quartz.layout.ts. Pin the clone to the last v4
release (v4.5.2) to restore known-good behavior.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Replace the Todoist-backed blumeops-tasks mise task with
`heph list --project Blumeops --json` (hephaestus, now at v1 prototype
on gilbert). Update task-discovery, rotation-reminder, and zk
references across docs; note the zk zettelkasten is migrating into
heph docs.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Mealie, Paperless, Immich, TeslaMate are now autodiscovered from their ringtail
Ingress gethomepage.dev annotations; the static services.yaml entries (from when
they were on minikube, which homepage-on-ringtail can't autodiscover) were
duplicating them.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- argocd: grant local break-glass admin the admin role (g, admin, role:admin);
previously only the Authentik admins group had access, locking out admin
once its token expired (policy.default is unset).
- alloy-k8s: repoint the teslamate blackbox probe from the deleted minikube
service to https://tesla.ops.eblu.me/ (Caddy over Tailscale), like immich.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Final step of the wave-1 indri-k8s migration. paperless, teslamate, mealie run on ringtail with data migrated, verified, and backed up (local + BorgBase offsite via PR #364).
- Remove minikube paperless/teslamate/mealie manifest dirs + ArgoCD app defs (prunes the parked Deployments/Services + redundant minikube mealie/paperless PVCs)
- Drop paperless/teslamate roles + ExternalSecrets from the minikube blumeops-pg cluster
- miniflux + authentik stay on minikube (later waves)
Finalization after merge: sync apps + databases to prune, then DROP DATABASE paperless/teslamate on indri's blumeops-pg (fresh safety dump taken first).
Reviewed-on: #365
Prereq for the wave-1 decommission. The cutover moved paperless+teslamate (postgres) and mealie (SQLite) to ringtail, but borgmatic and the Grafana TeslaMate datasource still pointed at the minikube copies — the migrated live data was unbacked since cutover, and dropping the minikube DBs would break the TeslaMate dashboards.
- Tailscale Service `blumeops-pg-ringtail` + Caddy L4 route `pg.ops.eblu.me:5434`
- borgmatic: teslamate + paperless postgres → :5434; mealie SQLite → ssh:eblume@ringtail
- Grafana TeslaMate datasource → pg.ops.eblu.me:5434
Deploy: sync databases-ringtail (tailscale svc) + grafana from branch; provision-indri --tags caddy,borgmatic; verify a backup run + dashboards. Unblocks the decommission PR.
Reviewed-on: #364
Post-merge rebuild of paperless/mealie/teslamate Nix images at the main
merge commit, replacing the feature-branch -nix tags. Image content is
identical; only the commit-sha suffix changes.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
GNU lives in the overhead — the X-Clacks-Overhead header — never on the
visible page. Keep the header, drop the footer.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A $29.60 Fly bill traced to ~1.25 TB/30d egress on forge.eblu.me (99.95% of
all proxy egress), ~71% of it AI scrapers (Meta meta-externalagent, OpenAI
GPTBot, Amazonbot, Bytespider) crawling the public mirror repos' infinite
git-history URL space and timing out Forgejo. robots.txt already disallowed
/mirrors/ but those agents ignore it, so enforce at the edge: return 403 (^~
to beat the regex asset locations), served as a roll-of-dishonour page with an
X-Naughty-Scrapers header. Mirrors stay reachable on the tailnet via
forge.ops.eblu.me. Tier 2 (UA denylist + Anubis) and the Cloudflare rejection
are documented in docs/explanation/ai-scraper-mitigation.md.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Image tags from PR #362 (v8.1.7-02859c5{,-nix}) referenced a branch
SHA that no longer exists on main after squash-merge. Rebuilt both
the dagger arm64 and nix amd64 variants from the squashed commit
(ecded30) and updated paperless + immich-ringtail to the new tags.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
## Summary
Weekly "make one non-local container local" pickup: immich-ringtail still pulled `docker.io/valkey/valkey:8.1.6` because the existing `containers/valkey/container.py` build was arm64-only.
- Adds `containers/valkey/default.nix` — nix-built amd64 valkey image, packaged by the ringtail nix-container-builder runner using `pkgs.dockerTools.buildLayeredImage`. Mirrors the existing `containers/authentik-redis/default.nix` pattern.
- `containers/valkey/container.py` keeps building the Alpine arm64 image for paperless on indri. Bumped both builds to upstream valkey 8.1.7 (Alpine 3.22 now ships `8.1.7-r0`; nixpkgs has 8.1.7).
- Splits `VERSION` (upstream app) from `ALPINE_PIN` (apk pin) in `container.py` so both build files can declare the same upstream version and pass `container-version-check`.
- Updates `service-versions.yaml`: current-version 8.1.7, refreshed last-reviewed, upstream-source now points at the canonical valkey-io releases page.
- Switches kustomizations:
- `immich-ringtail/kustomization.yaml`: `docker.io/valkey/valkey:8.1.6` → `registry.ops.eblu.me/blumeops/valkey:v8.1.7-02859c5-nix`, comment updated.
- `paperless/kustomization.yaml`: `v8.1.6-r0-fabca04` → `v8.1.7-02859c5`.
## Build
build-container run #563 — both jobs succeeded after a transient runner crash on the first dispatch (#562 build-nix), which surfaced two separate bugs that landed in a separate C0 on main:
- `runner-logs` silently returned 0 with no output when the log file didn't exist on indri
- `ssh indri` swallowing remote exit codes (fish login shell), which the wrapper now works around via a stdout marker
## Test plan
- [ ] `argocd app set immich-ringtail --revision valkey-nix && argocd app sync immich-ringtail`
- [ ] `argocd app set paperless --revision valkey-nix && argocd app sync paperless`
- [ ] Both valkey pods come Ready and start serving on :6379
- [ ] Immich app + paperless can read/write their respective cache
- [ ] After merge: rebuild from squashed main commit + update kustomization tags (squash-tag follow-up)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #362
`mise run runner-logs <run> -j <n>` previously silently succeeded with
no output when forgejo had no log for the task. Two layered causes:
1. zstdcat exits 0 even when the file is missing (writes "can't stat
… -- ignored" to stderr).
2. ssh to indri runs fish, which silently drops the remote exit code so
the subprocess returncode is always 0.
Probe `test -f` over SSH and parse a stdout marker (EXISTS / MISSING) to
detect the missing-log case, then report it explicitly with the indri
path and a hint about action_task.log_in_storage = 0 so the operator
knows where to look next.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Image was previously tagged with the unpoller-v3 branch SHA (1b27242),
which doesn't exist in main's history after squash-merge. Rebuilt from
the squashed commit so the tag references a reachable commit.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
## Summary
- Service Review pickup: unpoller (last reviewed 73 days ago).
- Upgrades unpoller from v2.34.0 to v3.2.0 (major version bump).
- Migrates the container build from a Dockerfile to a native Dagger pipeline (`containers/unpoller/container.py`) following the navidrome / miniflux pattern.
- Refreshes `service-versions.yaml` (last-reviewed, current-version).
## Breaking changes (upstream)
- **v3.0.0** — UniFi network API shifts (later 10.x). Some metric / event / log names and labels may have changed. Worth a follow-up sweep of the unpoller Grafana dashboard for missing series.
- **v3.2.0** — defaults to a 60s background poll feeding cached Prometheus scrapes (was on-demand poll per scrape). To restore previous behavior, set `interval = 0` in `up.conf`. Leaving the new default in this PR — every-15s scrapes will simply serve from cache, which is fine for our use.
## Build
- Image: `registry.ops.eblu.me/blumeops/unpoller:v3.2.0-1b27242`
- Built by build-container workflow run #559 from this branch.
## Test plan
- [ ] `argocd app set unpoller --revision unpoller-v3 && argocd app sync unpoller`
- [ ] Pod comes Ready
- [ ] Verify metrics exported (`Site/Client/UAP/USG/USW` counts in logs, `unpoller_*` series in Prometheus)
- [ ] Spot-check unpoller Grafana dashboard for missing series after the v3 API shift
- [ ] After merge: `argocd app set unpoller --revision main && argocd app sync unpoller`
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #361
Bluegreen kept timing out — the new green machine couldn't reach
"started" within Fly's 5-minute deploy budget. The cold-start sequence
(tailscaled → tailscale up → wait-for-MagicDNS → nginx startup) eats
most of that, leaving no headroom for healthcheck propagation.
For a single-machine proxy, bluegreen offers little benefit anyway:
no warm second instance, so trading 5-10s of downtime for predictable
completion is the right call.
Switch ringtail.yml from forge.eblu.me (Fly proxy, WAN) to
forge.ops.eblu.me (Caddy on indri, tailnet). Ringtail is always
on the tailnet — the WAN round-trip was overhead and made
provision-ringtail fail any time Fly was slow or down.
The OMEN 27i IPS pumps brightness when its refresh swings into the low
VRR range during low-framerate content (game cutscenes), producing a
~20Hz flicker that compounds over a session until a reboot. GPU health
is clean (no Xid/ECC/thermal); pinning fixed 165Hz eliminates it.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Wine/Proton game segfaults (e.g. Diablo IV) produced multi-GB cores that
systemd-coredump spent minutes compressing to disk, pinning the CPU and
freezing the desktop. Cap ProcessSizeMax/ExternalSizeMax at 1G (oversized
cores logged but skipped) and MaxUse at 2G to bound the store.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The verify step pointed to the main repo page, but the "Synchronize now"
button is in the Mirror settings section of the settings page.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The pbpaste | op item edit ... "field[password]=-" stdin syntax is
rejected by op 2.34 as "invalid JSON" — recent op versions treat
piped input as a full JSON template, not a single field value.
Procedure now uses an inline assignment via a local fish variable.
1Password's desktop app names exports as
1PasswordExport-<uuid>-<timestamp>.1pux automatically — you can't
choose the name. Procedure now points the task at that glob.
Added vault split (blumeops vs Personal), noted onepassword-connect
runs on both indri and ringtail, and lifted op CLI guidance from
agent memory into the card. Bumped last-reviewed.
## Summary
Removes the compensating-controls (CC) framework. Prowler and Kingfisher continue to run weekly and produce reports; the Prowler mutelist YAML files stay in place but no longer carry \`CC: <id>\` prefixes — each entry now just keeps a free-form \`Description\` of why it's muted.
The CC review cadence proved to be more process overhead than this single-operator homelab needed.
## What changed
**Deleted**
- \`compensating-controls.yaml\` — the CC registry
- \`mise-tasks/review-compensating-controls\` — the staleness-review task
- \`docs/how-to/operations/review-compensating-controls.md\`
- \`docs/how-to/operations/record-review-evidence.md\` (was aspirational)
- \`docs/explanation/compliance-mute-categories.md\` (proposed-future CC/NA/RA work)
- 5 orphan \`+review-cc-*\` / \`+compliance-mute-categories\` changelog fragments
**Modified**
- 6 mutelist YAML files: stripped \`CC: <id>.\` prefix from every \`Description\` / \`statement\` field, kept the free-form text
- \`mise-tasks/review-compliance-reports\`: removed CC mentions from docstrings, panel text, and the node-verification table title. Node-verification logic itself is unchanged.
- \`docs/reference/operations/security.md\`: removed the "Compensating controls" section
- \`docs/how-to/operations/read-compliance-reports.md\`: rewrote step 3 of "Acting on findings" to point at the mutelist YAML directly
- \`docs/changelog.d/prowler-iac-mutelist.infra.md\`: rewrote to drop the "two new compensating controls" framing
## What did not change
- All Prowler manifests (cronjobs, RBAC, PVs, kustomization) — scans still run on the same schedule
- The Kingfisher deployment
- The trivy-shim in the Prowler container — that's about Trivy ignorefile plumbing, independent of the CC concept
- The mutelist entries themselves — each \`Resources\` list is unchanged; only the prose of \`Description\` was edited
- \`CHANGELOG.md\` — historical releases are left as-is
## Test plan
- [ ] Wait for human review before deploying — once merged, re-point ArgoCD: \`argocd app set prowler --revision main && argocd app sync prowler\` (no manifest changes besides the ConfigMap, so impact is limited to muted-finding descriptions in next week's report)
- [ ] Confirm next weekly Prowler K8s CIS run (Sunday 3am) still completes and produces a report on sifaka
- [ ] Confirm next weekly Prowler IaC run still honors \`trivyignore.yaml\` (the trivy shim is untouched but the ignorefile content was rewritten)
- [ ] \`mise run review-compliance-reports\` — verify node-verification block still runs and prints the renamed table title
Reviewed-on: #359
Grafana uses an RWO PVC for SQLite + Bleve search index. RollingUpdate
spawns the new pod before terminating the old one, so the new pod
crashloops on the index lock until rollout timeout. Recreate terminates
the old pod first, letting the new pod acquire the lock cleanly.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Immich migrated to ringtail's k3s cluster but the probe still targeted
the in-cluster service DNS on indri's minikube, firing ServiceProbeFailure
indefinitely. Moved the target into alloy-ringtail's config so the probe
runs in the cluster where immich actually lives.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Image v1.1.3-3645098-nix was built directly on ringtail and pushed via
skopeo, bypassing the Forgejo runner: indri was severely overloaded
(load avg 24.92, minikube VM at 344% CPU) and the workflow-dispatch
endpoint timed out. The image content is identical to what the runner
would have produced — same default.nix at commit 3645098 (on main),
same NIX_PATH (current nixpkgs flake), same skopeo invocation. Tag
short-sha matches the commit that defines the recipe so we aren't
pinning to a ghost.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Wheel/sdist + FOD hashes probed on ringtail. Full nix-build verified
end-to-end before commit.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
UE5 writes Saved/running.dat as a "session in progress" marker. If
the previous session exited uncleanly (SIGKILL, crash), it lingers,
and SN2 pops up an invisible 0×0 Error dialog at next launch that
the GameThread blocks on forever — visible only as a black screen
with a spinning loader. Wrap the Steam command to clear the marker
files before each launch.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
PR #358 was squash-merged so the branch commit b8c7783 baked into the
prior image tag isn't reachable from main's history. Rebuild from main
HEAD (a33fa47) and retag. Image content is byte-identical (FOD is
content-addressed, inputs unchanged); only the SHA in the tag changes
so future provenance tracing stays on main.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
## Summary
Deploys `adelaide-baby-shower-app` **v1.1.2** to ringtail k3s.
- Bumps `containers/shower/default.nix` `version` to 1.1.2.
- Refreshes sdist + wheel `fetchurl` hashes against the forge PyPI artifacts.
- Re-probed FOD `outputHash` on the nix-container-builder runner (ringtail) and pinned the new closure hash.
- Bumps kustomize `newTag` to `v1.1.2-b8c7783-nix` (built from this branch's tip).
- Bumps `service-versions.yaml` entry for shower to `1.1.2` / `last-reviewed: 2026-05-15`.
## Build provenance
Built by Forgejo Actions run #553 on `nix-container-builder` (ringtail) at commit `b8c7783`. After merge a C0 follow-on will rebuild from main and retag so future provenance points at main history.
## Test plan
- [ ] `argocd app set shower --revision shower-v1.1.2 && argocd app sync shower` deploys cleanly
- [ ] Pod migrates the SQLite PV and serves at `shower.ops.eblu.me` / `shower.eblu.me`
- [ ] No new errors in pod logs after `collectstatic` + gunicorn boot
Reviewed-on: #358
Lets Subnautica 2 (and any other game) opt into the GE-Proton
build via Steam's per-game compatibility tool override, as a
workaround for the Proton Experimental + DXVK D3D12 Mercuna hang.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
## Summary
Nightly borgmatic backups have been failing for 2 days. Root cause: the
shower SQLite dump `before_backup` hook (added in PR #349) referenced
`kubectl --context=k3s-ringtail`, but indri's kubeconfig deliberately
doesn't carry the ringtail credentials. The hook's failure aborted the
entire run, taking out *both* the local sifaka repo and the BorgBase
offsite. Verified the last good archive was `indri-2026-05-11T02:00`.
## Approach
ssh into ringtail and run `k3s kubectl` there — no indri-side
kubeconfig needed. `/etc/rancher/k3s/k3s.yaml` is mode 644 so no sudo
required, and the existing ssh access from indri to ringtail works.
Inline-shell quoting got hairy fast (fish on ringtail rejected `POD=...`
bash syntax; the nix shower image lacks `tar` so `kubectl cp` fails).
Pulled the dump logic into `~/bin/borgmatic-k8s-sqlite-dump`, deployed
by the ansible role. Each dump entry now declares a `target`:
- `local:<context>` — local kubectl with explicit context (mealie)
- `ssh:<user@host>` — ssh + `k3s kubectl` on the cluster host (shower)
Bytes come back via `kubectl exec ... -- cat` instead of `kubectl cp`
since `cp` needs `tar` in the pod (nix-built containers don't bundle it).
## Test plan
- [x] `mise run provision-indri -- --tags borgmatic --check --diff` shows expected diff
- [x] Apply, helper script deployed at `~/bin/borgmatic-k8s-sqlite-dump`
- [x] Helper invoked directly with `ssh:eblume@ringtail` produces a valid 288 KB SQLite file
- [x] Full `borgmatic create` completes without errors — both mealie.db (1.7 MB) and shower.db (288 KB) appear in `~/.local/share/borgmatic/k8s-dumps/`, archive `indri-2026-05-13T17:31:02` written to sifaka borg repo
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #357
## Summary
C2 Mikado chain to move the entire Immich stack (server, ML, valkey,
postgres) off `minikube-indri` and onto `k3s-ringtail`. Immich is the
largest single tenant on minikube (~1.5 GiB resident) and minikube is
currently memory-saturated (97% RAM, swapping). This is the first
concrete chain in the broader indri-k8s decommission effort.
This PR contains the planning layer only — 7 cards (1 goal + 6
prerequisites). Implementation cycles follow per the Mikado Branch
Invariant.
## Goal end-state
- Immich `server`, `machine-learning`, `valkey` on ringtail.
- ML pod uses ringtail's RTX 4080 (performance win — currently
CPU-only).
- CNPG `immich-pg` (PG17 + VectorChord) runs on ringtail.
- Library still on sifaka NFS — ringtail mounts the same path.
- `photos.ops.eblu.me` reroutes through Caddy → ringtail ingress.
- Minikube `immich` and `immich-pg` are removed.
## Cards
| Card | Depends on |
|---|---|
| `migrate-immich-to-ringtail` (goal) | all six below |
| `cnpg-on-ringtail` | — |
| `immich-pg-on-ringtail` | cnpg-on-ringtail |
| `immich-pg-data-migration` | immich-pg-on-ringtail |
| `sifaka-nfs-from-ringtail` | — |
| `immich-app-on-ringtail` | immich-pg-on-ringtail, sifaka-nfs-from-ringtail |
| `immich-cutover-and-decommission` | immich-pg-data-migration, immich-app-on-ringtail |
## Key constraints
- **No data loss.** Downtime is acceptable; data loss is not. Two
surfaces matter: postgres (ML embeddings, face data — slow to
re-derive) and the library files (don't move, but NFS access from
ringtail must be verified).
- **Migration method:** Option A is a CNPG `externalCluster`
basebackup → promote. Option B is `pg_dump`/`pg_restore` as a
documented fallback. Either way, dry-run against a scratch
cluster first.
- **Why pg moves too** (not cross-cluster): keeping pg on minikube
would block the whole decommission, and Immich is chatty with pg
so tailnet round-trips would hurt.
## Test plan
- [ ] Plan review — does the dependency graph make sense?
- [ ] `mise run docs-mikado migrate-immich-to-ringtail` shows the
chain correctly.
- [ ] Per-card implementation cycles land separately (commit
convention enforced by hook).
Reviewed-on: #356
After the static IP change, k3s/flannel pod networking broke because
ip_forward was 0. NixOS doesn't enable IP forwarding by default — it
was previously being set implicitly somewhere in the NM-managed /
scripted-DHCP path. With static networking we have to set it ourselves.
Verified at runtime via sysctl -w before adding here; pod outbound
came back immediately and Tailscale VIP services recovered without
any pod restarts.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Removes DHCP lease renewal as a failure mode on ringtail after an outage
on 2026-05-12 where the IP and routes silently disappeared from enp5s0
without any kernel link event. NetworkManager stays enabled for wireless
fallback but no longer manages the wired interface.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
PR #354 was squash-merged so the branch commit 444ff91 baked into the
prior image tag isn't reachable from main's history. Rebuild from main
HEAD (3c7967e) and retag. Image content is byte-identical (FOD is
content-addressed, inputs unchanged); only the SHA in the tag changes
so future provenance tracing stays on main.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
## Summary
Deploys `adelaide-baby-shower-app` **v1.1.0** to ringtail k3s.
### App changes (since v1.0.2)
- **Four-phase `ShowerState`** replaces the boolean `locked` flag — `pre_event` → `party` → `prizes_locked` → `event_locked` — with a backfill migration that maps `locked=True → pre_event`, `locked=False → party`.
- **Guest memories**: append-only photos + comments panel where guests can leave notes for the baby. Adds `GuestPhoto` + `GuestComment` models with file-extension validators and a max-size validator; new `shower.imaging` module for thumbnail generation.
- **Admin + QR polish**: configurable host link, fixed "View Site" URL, guest-facing QR copy improvements, contest tweaks.
Three Django migrations run automatically in the entrypoint against the SQLite PV:
- `0009_shower_phase`
- `0010_guest_memories`
- `0011_book_description`
No ConfigMap / env-var changes. The deploy uses `strategy: Recreate` with a single replica, so the old pod releases the data PVC before the new one mounts it and runs migrations.
### Container build changes
The v1.1.0 tag exposed a latent issue with the Forgejo PyPI install path:
- The recent commit [2d38418e](2d38418e) closed the forge package leak at the Fly edge by blocking `/api/packages/*` publicly.
- Forgejo's PyPI simple index returns absolute file URLs hardcoded to its public `ROOT_URL` (`forge.eblu.me`), so pip-installing from the tailnet index URL still tries to download from `forge.eblu.me` → 403.
- Previous shower builds escaped this because their FOD outputs were already in the nix store; bumping to a new version forced a fresh pip run that hit the block.
Fix mirrors what we already do for the sdist: both wheel and sdist are pulled via direct `fetchurl` against `forge.ops.eblu.me`, then the wheel is copied to TMPDIR under its clean filename (nix store path's hash prefix breaks pip's wheel-filename parser) and handed to pip as a local path. The forge `--extra-index-url` is no longer needed.
FOD outputHash pinned to `sha256-kTNOswobtkgyQmmqbQM8XO4vvaGg57nCuuZGbNXb0NM=` from run 547. Image: `registry.ops.eblu.me/blumeops/shower:v1.1.0-444ff91-nix`.
### Adjacent finding (already handled)
The ringtail `gitea-runner-nix_container_builder` systemd unit was left `inactive` after the recent `provision-ringtail` (matches the known `sshd-restart-hangs-mux` lesson — the rebuild changed the unit's PATH closure + config.yaml, systemd stopped it, then the playbook hung before the activation could restart it). Manually started; the existing memory `lesson_provision_ringtail_ssh_hang.md` was extended to mention the runner as the canary service to check after provisions.
## Test plan
- [ ] `argocd app diff shower --revision shower-v1.1.0` — review the manifest change
- [ ] `argocd app set shower --revision shower-v1.1.0 && argocd app sync shower`
- [ ] `kubectl --context=k3s-ringtail logs -n shower deploy/shower` — confirm migrations 0009/0010/0011 applied, no errors
- [ ] Hit `https://shower.ops.eblu.me/` (tailnet) — splash page renders, phase indicator visible
- [ ] Hit `https://shower.ops.eblu.me/host/` — host console loads, phase dropdown shows the four states
- [ ] Hit `https://shower.eblu.me/` (public via Fly) — splash page still served
- [ ] After merge: `argocd app set shower --revision main && argocd app sync shower`
Reviewed-on: #354
Transient lock file written by the ScheduleWakeup harness tool when
Claude paces its own work between long-running operations. Not config,
not state worth checking in.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
## Summary
- Recurring weekly lockfile refresh for `nixos/ringtail/flake.lock`.
- Inputs updated: `disko`, `home-manager`, `nixpkgs`.
- `nixpkgs-services` was deliberately skipped (per overlay convention — pinned services bump only on intentional update).
- Generated via `dagger call flake-update --src=. --flake-path=nixos/ringtail`.
## Test plan
- [x] `prek` hooks pass
- [ ] After merge: `mise run provision-ringtail` to deploy
- [ ] Then check for kernel update per [[manage-lockfile]]
## Notes
- Not deployed from this PR — provisioning is a follow-up.
Reviewed-on: #352
## Summary
- Recurring service review for `mealie`.
- Upstream is at **v3.17.0** (released 2026-05-06); deployed image is **v3.12.0** — 5 minor versions behind.
- Container is built locally from the forge mirror (`containers/mealie/Dockerfile`), so upgrade requires a fresh build + changelog review for breaking changes between v3.12 and v3.17.
- Deferring the actual upgrade to a separate task; this PR just refreshes `last-reviewed` and captures the gap in `notes`.
## Test plan
- [x] `prek` hooks pass
- [ ] Follow-up: open task to bump `containers/mealie/Dockerfile` `CONTAINER_APP_VERSION`, build, and update kustomization tag
## Notes
- No deployment changes in this PR.
Reviewed-on: #351
## Summary
- Recurring compensating-control review (oldest stale control: 42 days).
- Verified the control is in effect on both clusters:
- `alloy-k8s` on minikube-indri — Synced/Healthy, DaemonSet 1/1 ready
- `alloy-ringtail` on k3s-ringtail — Synced/Healthy
- `loki` (`monitoring/loki-0`) — Running, receiving logs (52 restarts in 18h is worth watching but not blocking review)
- Generalized the description: previously named only minikube, but the indri→ringtail migration means we now operate two clusters and both rely on this control.
- Added a follow-up note: enabling native apiserver audit logging is far more tractable on k3s (`--audit-log-path` / `--audit-policy-file`) than it was on minikube — worth revisiting once the migration concludes.
## Test plan
- [x] `prek` hooks pass
- [x] Verified alloy + loki status via `kubectl --context=minikube-indri` and `argocd app get`
## Notes
- No deployment changes.
Reviewed-on: #353
Standard squash-merge dance per
docs/how-to/deployment/build-container-image.md#Squash-merge-and-container-tags
— retags from v1.0.2-039d9b9-nix (branch SHA) to v1.0.2-292d354-nix
([main] tag from run 544 built off the merge commit). Functionally
identical; preserves source traceability.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
## Summary
Brings up the Adelaide / Heidi / Addie baby shower app on ringtail k3s with the public/private split that the app's hosting contract calls for: `shower.eblu.me` (public, via Fly proxy) and `shower.ops.eblu.me` (tailnet). App is consumed as a wheel from the Forgejo PyPI index — source lives at [`adelaide-baby-shower-app`](https://forge.eblu.me/eblume/adelaide-baby-shower-app).
### What's included
- **ArgoCD app + manifests** under `argocd/manifests/shower/` (deployment, service, ProxyGroup ingress, ConfigMap for `DJANGO_DEBUG`/`DJANGO_ADMIN_URL`, ExternalSecret for `DJANGO_SECRET_KEY` from 1Password item `Shower (blumeops)`, NFS PV on sifaka, RWX media PVC, RWO local-path data PVC for SQLite). Recreate rollout because SQLite is single-writer.
- **Public surface** (`fly/`): new `shower.eblu.me` server block proxying to `shower.ops.eblu.me`. `/admin/` returns 403 at the edge except `/admin/login/` and `/admin/logout/`, which are rate-limited via a new `shower_auth` zone. `X-Clacks-Overhead` on. GNU Terry Pratchett.
- **fail2ban** filter (`shower-admin-login.conf`) matching 401/403/429 on `/admin/login/` and jail (`shower.conf`) with `maxretry=5/findtime=600/bantime=3600`. The `nginx-deny` action was generalized to take a per-jail `nginx_deny_file` so the shower has its own deny list (forge keeps using the legacy default).
- **Caddy** route on indri (`shower.ops.eblu.me` → `https://shower.tail8d86e.ts.net`).
- **Pulumi** Gandi CNAME `shower.eblu.me → blumeops-proxy.fly.dev.`.
- **Grafana** APM dashboard `configmap-shower-apm.yaml` (request rate, error rate, failed admin login count, latency percentiles, bandwidth, access logs) mirroring `docs-apm.json` with a `host="shower.eblu.me"` filter.
- **Container** `containers/shower/default.nix` — `dockerTools.buildLayeredImage` with a nixpkgs Python and a startup wrapper that creates `/app/data/.venv`, pip-installs `adelaide-baby-shower-app==1.0.0` from the forge PyPI index on first boot, runs migrations + collectstatic, and execs gunicorn. A `local_settings.py` shim pins `DATABASES.NAME`/`MEDIA_ROOT`/`STATIC_ROOT` to absolute paths so they don't end up in site-packages.
- **Docs** runbook at `docs/how-to/operations/shower-app.md` linked from the apps registry, plus changelog fragments.
### Defense layers on the public surface
1. fly nginx geo+fail2ban `$shower_banned` (per-service deny list)
2. fly nginx `limit_req zone=shower_auth` (3 r/s per Fly-Client-IP)
3. django-axes (5 fails / 1h, keyed on username+ip_address)
4. edge `/admin/` block (returns 403 for anything that isn't login/logout)
## Prerequisites for the user to do (NOT in this PR)
Halted on these per request — they touch shared/manual systems:
- [x] **NFS share** on sifaka: `/volume1/shower`, NFS rule for ringtail RW, `chown 1000:1000`
- [ ] **1Password item** `Shower (blumeops)` in the blumeops vault with a freshly minted `secret-key` field (`openssl rand -base64 48`) — do NOT reuse anything that has lived in git
- [ ] **Container build**: `mise run container-build-and-release shower`, then update `images[].newTag` in `argocd/manifests/shower/kustomization.yaml` to the resulting `v1.0.0-<sha>-nix`
- [x] **DNS**: `mise run dns-up` after merge
- [x] **Fly cert**: `fly certs add shower.eblu.me -a blumeops-proxy`
- [ ] **Caddy push**: `mise run provision-indri -- --tags caddy`
- [ ] **Fly redeploy** to pick up the new nginx block + fail2ban jail: `mise run fly-deploy`
- [ ] **ArgoCD sync**: `argocd app set shower --revision shower-app-deploy && argocd app sync shower` to test from this branch before merging
## Test plan
- [ ] Container builds successfully on nix-container-builder runner
- [ ] Pod starts, migrations run, gunicorn answers on :8000
- [ ] `kubectl --context=k3s-ringtail -n shower logs deploy/shower` clean
- [ ] `curl -sf https://shower.ops.eblu.me/` returns the splash page (tailnet)
- [ ] `curl -I -H "Host: shower.eblu.me" https://blumeops-proxy.fly.dev/` returns 200 (pre-DNS verification)
- [ ] `curl -I -H "Host: shower.eblu.me" https://blumeops-proxy.fly.dev/admin/users/` returns 403 (edge block)
- [ ] `curl -I -H "Host: shower.eblu.me" https://blumeops-proxy.fly.dev/admin/login/` returns a Django login response
- [ ] After DNS is up: `curl -I https://shower.eblu.me/` returns 200 with `X-Clacks-Overhead`
- [ ] Grafana dashboard "Shower APM" appears and starts showing traffic
- [ ] `mise run services-check` passes
Reviewed-on: #349
The previous Dockerfile chowned /app/config to 1000:1000 so the runtime
user could seed missing skeleton configs (e.g. proxmox.yaml) and write
/app/config/logs. The nix derivation didn't replicate that, so the new
amd64 image crashed with EACCES on cold start (fixed-forward — caught
during ringtail cutover, ArgoCD #348).
Add fakeRootCommands to dockerTools to create /app and /app/config and
chown them at build time. The deployment's ConfigMap subPath mounts
leave the parent directory as image filesystem, so its ownership has to
be set at build time, not at runtime.
Repoint the ArgoCD Application destination from minikube to ringtail and
bump the image tag to the new amd64 nix-built v1.11.0-b87f62e-nix.
Rework services.yaml for the autodiscovery shift: 11 services that
previously auto-populated via minikube Ingress annotations (ArgoCD,
Immich, Kiwix, Mealie, Miniflux, Grafana, Prometheus, Navidrome,
Paperless, TeslaMate, Transmission) become explicit static entries with
their widget configs preserved. Conversely, the ringtail services that
will now auto-populate (Frigate/NVR, Authentik, Ntfy) are removed from
the static list to avoid duplicates; Ollama becomes newly visible.
Add a Content group for Immich/Kiwix/Miniflux which previously lived
under the autodiscovered "Content" group from annotations.
Replace Dockerfile (arm64-only, indri-built) with a nix derivation
adapted from nixpkgs pkgs/by-name/ho/homepage-dashboard. Built via the
nix-container-builder runner on ringtail, producing an amd64 image
suitable for k3s.
Includes the upstream Next.js file-system-cache patch to avoid
prerender cache write failures on a read-only nix store path
(nixpkgs issues #328621 and #458494).
Pinned to v1.11.0 (current production version).
Routine post-squash-merge cleanup. Bumps the ProxyClass image tag from
the now-orphaned PR branch SHA (67af7a8) to the merge commit SHA
(0108b68) so the deployed image stays traceable after branch cleanup.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
## Summary
Adds the first cut of a local nix build for `docker.io/tailscale/tailscale` and rewires only the ringtail tailscale-operator overlay to use it. Indri's overlay continues pulling upstream — minikube on indri is being decommissioned in favor of ringtail's k3s, so investing in dual-cluster routing here would be wasted churn.
## Changes
- `containers/tailscale/default.nix` — `buildGoModule` over `cmd/tailscale`, `cmd/tailscaled`, `cmd/containerboot`; packaged via `dockerTools.buildLayeredImage` with `cacert`, `iptables` (legacy symlink to match upstream Synology compat), `iproute2`, `tzdata`, `busybox`.
- `argocd/manifests/tailscale-operator-ringtail/kustomization.yaml` — kustomize `images:` rewrite swapping `docker.io/tailscale/tailscale` → `registry.ops.eblu.me/blumeops/tailscale:v1.94.2-67af7a8-nix`.
- `docs/changelog.d/mirror-tailscale-container.infra.md` — fragment.
## Pin rationale
v1.94.2 matches `service-versions.yaml:96` and the current ProxyClass exactly — this PR is "make it local," not "upgrade tailscale." Version bumps come as follow-up C0/C1 changes once we decide to test newer (v1.96.x had a Fly-side MagicDNS regression; v1.98.0 is current upstream stable).
## Test plan
- [x] Image built successfully on ringtail nix-container-builder (run #528).
- [x] Image visible in registry: `registry.ops.eblu.me/blumeops/tailscale:v1.94.2-67af7a8-nix`.
- [ ] Deploy from branch: `argocd app set tailscale-operator-ringtail --revision mirror-tailscale-container && argocd app sync tailscale-operator-ringtail`.
- [ ] Verify proxy pods restart with new image and existing tailnet ingresses (e.g., authentik, immich, tempo) keep resolving.
- [ ] After merge: rebuild on main SHA, update kustomization, run `services-check`.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #347
Runtime grafana pod matches the manifest and the CC's claim; bumped
last-reviewed. Noted that retiring init-chown-data in favor of fsGroup
alone should wait until grafana migrates to ringtail's k3s, since the
storage backend will change.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Captures the CC vs NA vs RA distinction surfaced during the 2026-05-03
weekly compliance review (CVE-2026-31789), and the image-scan mutelist
gap that blocks acting on it. Links the new article from the
review-compensating-controls how-to so it isn't orphaned.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Security fixes only (TLS verification on metrics client, CORS
Allow-Credentials suppression on wildcard origin, manifest/API-key
body-size limits, dependabot bumps). No config changes required;
re-built from source on indri and bounced launchagent.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Combine mint+store into a single command with both fish and bash
forms (the doc previously only showed manual paste). Document the
1Password CLI "Password item requires ps value" validator error and
the placeholder-password workaround for Password-category items with
empty primary password fields.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Routine post-merge follow-up after #346. Branch SHA tag (946fa75) replaced
with the main-SHA-built tag (fabca04) so paperless and immich reference an
image traceable to a commit on main.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
## Summary
- Add native Dagger build of valkey 8.1.6-r0 on Alpine 3.22 at `containers/valkey/`
- Swap paperless redis sidecar and immich-valkey from `docker.io/valkey/valkey:8.1-alpine` to `registry.ops.eblu.me/blumeops/valkey:v8.1.6-r0-946fa75`
- Resolves the DR-2026-04 TODO in paperless kustomization about multi-arch redis
## Why
Move toward fully locally-built containers for supply chain control. Paperless and immich both pulled the same upstream tag — one mirror serves both. Authentik's nix-built Redis stays separate (different image entirely).
## Risk
Low. Both sidecars are stateless caches:
- paperless redis: no volumeMount (in-pod localhost, pure memory)
- immich-valkey: `emptyDir` (cache only)
Pod restart rebuilds the cache. Smoke-tested locally (PING/SET/GET roundtrip on `valkey 8.1.6` with `--bind 0.0.0.0 --protected-mode no`).
## Test plan
- [ ] After merge: `mise run container-build-and-release valkey` to rebuild with main SHA
- [ ] Update kustomizations to the `[main]` SHA tag (C0 follow-up)
- [ ] `argocd app sync paperless` and `argocd app sync immich`
- [ ] Verify pods come up healthy (paperless OCR queue functional, immich job queue functional)
- [ ] `mise run services-check`
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #346
Verified Forgejo runner is registered only to forge.ops.eblu.me and the
forge has registration disabled, so no untrusted users can trigger
privileged CI. Tightened notes to reflect the closed-forge mechanism
(not a per-repo allow-list).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Bumped documented image tag to 0.20.4 (matches kustomization newTag),
added the two qwen3.5 models from models.txt, and stamped the card.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Completes the v1.16.0 fleet upgrade for the fourth alloy service
(type: ansible, built from source on indri). Binary built on gilbert
with Go 1.26.2 + CGO, scp'd to indri, codesigned, LaunchAgent reloaded.
Service reports clean WAL replay and resumed metric/log shipping.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Per the build-container-image squash-merge convention, rebuild alloy v1.16.0
container images from the main SHA (9564435) and update the three alloy
kustomizations to reference :v1.16.0-9564435[-nix] instead of the branch
SHA :v1.16.0-26a3ab5[-nix] left over from #345.
Both images were rebuilt locally on gilbert (dagger) and ringtail (nix)
because indri is still under heavy macOS memory-compressor pressure (see
separate ticket); CI on indri can't reliably run the dagger publish step.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Bump Grafana Alloy v1.14.0 → v1.16.0 across all four services (alloy-k8s, alloy-ringtail, alloy-tracing-ringtail; alloy native ansible). Also migrate the indri build path from `Dockerfile` to a native Dagger `container.py` per the build-container-image migration playbook.
## Highlights from upstream
- v1.15: database observability promoted to stable, OTel Collector → v0.147.0
- v1.16: clustering for `loki.source.kubernetes_events`, MySQL exporter 0.19.0
- One pre-existing breaking change in v1.15 (`loki.source.awsfirehose` undocumented metric prefix rename) — not used here.
## Build infra
Alloy v1.16.0's go.mod requires Go 1.26.2. The nix derivation now uses `pkgs.go_1_26` with `GOTOOLCHAIN=local` to avoid auto-downloading a toolchain blob that violated the fixed-output rule.
## Test plan
- [ ] CI: `mise run container-build-and-release alloy --ref alloy-v1.16.0` (dispatched as run 522; nix job to be re-triggered with the v1.16.0 goModules outputHash once the local ringtail build surfaces it)
- [ ] After CI green, bump `images[].newTag` in three kustomizations to the new `-<sha>` and `-<sha>-nix` tags, deploy from this branch via `argocd app set <app> --revision alloy-v1.16.0 && argocd app sync <app>`
- [ ] Manual rebuild of macOS native binary on gilbert (per ansible/roles/alloy README) and `mise run provision-indri -- --tags alloy --check --diff`
- [ ] `mise run services-check` after merge & redeploy
Reviewed-on: #345
## Summary
Monthly tooling dependency refresh, with a one-time conversion from version-tag pins (`rev = "vX.Y.Z"`, `image:tag`, `>=`) to SHA / digest pins everywhere.
## Changes
- **prek hooks**: all `rev = "vX.Y.Z"` → commit SHA + `# vX.Y.Z` comment. Bumped trufflehog (3.94.0→3.95.2), kingfisher (1.91.0→1.97.0), ruff (0.15.7→0.15.12), shfmt (3.13.0→3.13.1), prettier (3.8.1→3.8.3), actionlint (1.7.11→1.7.12).
- **fly/Dockerfile**: tag pins → `image@sha256:...` digest pins. Bumped nginx (1.29.6→1.30.0-alpine), tailscale (v1.94.1→v1.94.2 — still inside the safe pre-1.96.5 range), alloy (v1.14.1→v1.16.0).
- **mise-tasks**: PEP 723 inline deps converted from `>=` to `==` (PEP 508 doesn't support hashes inline). All scripts pinned to current latest: rich 15.0.0, typer 0.25.0, pyyaml 6.0.3, httpx 0.28.1.
- **prek `additional_dependencies`**: ansible-lint==26.4.0, ansible-core==2.20.5.
- **taplo-lint**: pass `--no-schema`. Upstream's `--default-schema-catalogs` returns a format taplo v0.9.3 can't parse — we don't validate against TOML schemas anyway, so this turns off the broken catalog fetch.
- **docs/update-tooling-dependencies**: documents the SHA-pin convention, `docker buildx imagetools inspect` for digest lookup, and `prek clean` before re-verifying (cache grows to several GiB).
Forgejo workflow `actions/checkout@v6.0.2` was already at the latest SHA — no change.
## Test plan
- [x] `prek run --all-files` passes after `prek clean`
- [x] `deploy-fly` workflow builds and deploys the new fly image on merge
- [x] `fly status -a blumeops-proxy` healthy after deploy
- [x] Spot-check a few mise tasks (`mise run blumeops-tasks`, `mise run docs-check-links`) to confirm pinned deps resolve cleanly
Reviewed-on: #344
## Summary
Follow-up to #342. The cv and docs services are now live on indri (Caddy file_server backed by ansible-managed tarball extraction) and verified working. This PR removes the dead minikube artifacts and the tooling shims that referenced them.
## Changes
**Deletions:**
- ``argocd/apps/{cv,docs}.yaml``
- ``argocd/manifests/{cv,docs}/`` (deployment, service, ingress, pdb, kustomization)
- ``containers/{cv,quartz}/`` (Dockerfiles + start scripts)
**Tooling:**
- ``mise-tasks/container-version-check``: remove the ``quartz``→``docs`` CONTAINER_TO_SERVICE mapping (containers/quartz no longer exists)
- ``service-versions.yaml``: bump ``docs.current-version`` to ``v1.16.0`` (the blumeops docs release tag) and trim the migration-window comment
## Live state context
The argocd Applications ``cv`` and ``docs`` were already deleted from the cluster manually as part of the cutover; this PR just removes the YAML files that the ``apps`` app-of-apps was still ingesting. After merge, ``argocd app sync apps`` will reconcile and the ``apps`` Application returns to Synced.
The Caddyfile ``handle_errors`` bug that briefly crashed all ``*.ops.eblu.me`` services during cutover is fixed in a separate C0 (``2ee53fe``) on main, not here.
## Test plan
- [x] ``mise run container-version-check --all-files`` clean
- [x] ``mise run service-review --type ansible`` shows cv at 1.0.3, docs at v1.16.0
- [ ] After merge: ``argocd app sync apps`` returns clean (cv/docs entries gone, no children to reconcile)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #343
The kind=static branch added in #342 put handle_errors inside the
@host handle{} block. handle_errors is a top-level site-block directive,
not an ordered HTTP handler, so Caddy refuses to load the config:
parsing caddyfile tokens for 'handle': directive 'handle_errors'
is not an ordered HTTP handler
This crash-loops the whole reverse proxy and takes down every
*.ops.eblu.me service. Tripped today during the live cv/docs cutover.
Fix: drop handle_errors and append /404.html as the final try_files
candidate. The 404 page is served with status 200 instead of 404, but
that's acceptable for a human-facing curated 404 — the page renders
correctly. Documented inline.
The running Caddy on indri already has the fixed config (deployed
manually during the cutover); this lands the fix in main so future
provision-indri --tags caddy runs don't re-break it.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
## Summary
Replace the cv (`cv.eblu.me`) and docs (`docs.eblu.me`) minikube Deployments with indri-native ansible roles. Caddy serves the extracted release tarballs directly via a new `kind: static` service-block — no daemon, no nginx pod, no ProxyGroup ingress on the request path. Mirrors the rationale of the recent devpi migration; part of the broader minikube wind-down.
## What's in this commit
- `ansible/roles/{cv,docs}` — sentinel-gated tarball download + extract into `~/{cv,docs}/content/`
- `ansible/roles/caddy/` — new `kind: static` branch in the Caddyfile template (encoded gzip, immutable cache headers for fingerprinted assets, optional `try_html` for Quartz-style clean URLs, optional per-path `download_paths` for the resume PDF's `Content-Disposition`)
- `ansible/playbooks/indri.yml` — wires `cv` and `docs` roles before `caddy`
- `service-versions.yaml` — both services flip to `type: ansible`. `docs.current-version` stays at `1.28.2` for this commit so `container-version-check` keeps passing while `containers/quartz/Dockerfile` still exists; it moves to the docs release tag in the cleanup commit
- `.forgejo/workflows/{cv-deploy,build-blumeops}.yaml` — deploy step now bumps `cv_version`/`docs_version` in the role defaults and pushes; running ansible + purging the Fly cache is manual from gilbert (matches devpi)
- Docs: `docs/how-to/operations/{cv,docs}-on-indri.md`, updated `docs/reference/services/{cv,docs}.md`, changelog fragment
## What is not in this commit
The dead artifacts. After PR review and successful cutover, a follow-up commit deletes:
- `argocd/apps/{cv,docs}.yaml` and `argocd/manifests/{cv,docs}/`
- `containers/cv/`, `containers/quartz/`
- `CONTAINER_TO_SERVICE['quartz']` mapping in `mise-tasks/container-version-check`
- bumps `docs.current-version` in `service-versions.yaml` to the release tag
## Cutover plan (manual, from gilbert, after review)
1. **Take down old:**
- Remove the cv and docs Applications: `argocd app delete cv --cascade && argocd app delete docs --cascade`
- Verify k8s namespaces gone: `kubectl --context=minikube-indri get ns | grep -E '^(cv|docs)\\b'` (should be empty)
- Verify tailnet MagicDNS no longer advertises the VIPs: `nslookup cv.tail8d86e.ts.net` and `nslookup docs.tail8d86e.ts.net` should both fail
2. **Bring up new:**
- `mise run provision-indri -- --tags cv,docs,caddy --check --diff` (already validated on branch)
- `mise run provision-indri -- --tags cv,docs,caddy`
- `fly ssh console -a blumeops-proxy -C "sh -c 'rm -rf /tmp/cache && nginx -s reload'"`
3. **Verify:** `mise run services-check` and the curl checks listed in `docs/how-to/operations/{cv,docs}-on-indri.md`
4. **Cleanup commit + merge.**
Total expected downtime: minutes (not the few-hour budget you authorized).
## Test plan
- [ ] `mise run provision-indri -- --tags cv,docs --check --diff` clean
- [ ] `mise run provision-indri -- --tags caddy --check --diff` shows only the cv + docs blocks changing as previewed in the PR thread
- [ ] After cutover: `cv.eblu.me`, `cv.ops.eblu.me`, `docs.eblu.me`, `docs.ops.eblu.me` all return 200
- [ ] `cv.eblu.me/resume.pdf` includes `Content-Disposition: attachment`
- [ ] A clean Quartz URL (e.g. `docs.eblu.me/explanation/agent-change-process`) resolves to the right page
- [ ] `mise run services-check` clean
- [ ] `mise run service-review --type ansible` shows cv and docs
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #342
Devpi now runs natively on indri (uv venv via ansible role), so the
Dagger container build at containers/devpi/ is unused. Removing it.
Also updated dagger.md examples to use 'miniflux' as the example
container-name argument.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
## Summary
Devpi was crash-looping under memory pressure on the minikube StatefulSet, breaking the Python toolchain across the repo (`mise run docs-mikado`, `prek`, every `uv pip install`). It moves to indri as a native LaunchAgent.
## What changed
- **New ansible role** `ansible/roles/devpi/`: installs `devpi-server` + `devpi-web` into a uv-managed venv, initializes the server-dir on first run via 1Password root password, runs as a LaunchAgent (`mcquack.eblume.devpi`) bound to `127.0.0.1:3141`. Bootstraps from upstream PyPI (so devpi can install itself on a fresh box).
- **Caddy**: `pypi.ops.eblu.me` now proxies to `http://localhost:3141`.
- **Playbook**: `indri.yml` gains pre_tasks for the root password and the new role.
- **service-versions.yaml**: devpi flipped from `type: argocd` to `type: ansible`.
- **ArgoCD**: removed `apps/devpi.yaml` and `manifests/devpi/`. The in-cluster Application, namespace, and PVC have been deleted.
- **Docs**: new how-to `docs/how-to/operations/devpi-on-indri.md`; `restart-indri.md` lists devpi in the LaunchAgent stop list.
## Already deployed (live on indri)
- Service running: `launchctl list mcquack.eblume.devpi` → PID 53888
- `curl https://pypi.ops.eblu.me/+api` returns 200 ✅
- `mise run docs-mikado` works again ✅
- 1.0G of cached PyPI data was migrated from the PVC to `~erichblume/devpi/server-dir/`
- Minikube namespace and PVC fully reclaimed
## Test plan
- [ ] `mise run services-check` (after merge)
- [ ] CI workflows that use devpi succeed
- [ ] No regressions in tools that depend on `pypi.ops.eblu.me` (prek, uv-script tasks, dagger pipelines)
## Context
This is the C1 prelude to a planned C2 chain (`mikado/retire-minikube-indri`) to retire minikube on indri entirely. Doing devpi as a standalone C1 was the right call because (a) it was urgent — it was breaking the toolchain — and (b) it shakes out the migration recipe before we commit to a multi-leaf chain.
Reviewed-on: #341
Verified TTL=604800s and hostPID limited to ephemeral Prowler CronJob
on indri. Noted that alloy-tracing on ringtail also uses hostPID but
is out of scope until Prowler scans ringtail (tracked in Todoist).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Marked last-reviewed: 2026-04-29. Fixed the storage layout table —
`/config/` is an emptyDir (ephemeral), not NFS, and the watch directory
is disabled. Documented the transmission-exporter sidecar that exposes
Prometheus metrics on port 19091.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Squash-merge of #340 changed the SHA. Bump prowler tag from
v5.23.0-2daf629 (PR branch) to v5.23.0-495e45d (main HEAD) so the
Dockerfile changes are present in the image deployed off main.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
## Summary
The weekly Prowler IaC scan reported 6 critical findings against `argocd/manifests/`. They split cleanly into two patterns:
- **Legitimate-by-design RBAC → mute with new compensating controls**
- `external-secrets-controller`, `external-secrets-cert-controller` manage `secrets` (KSV-0041) and the cert-controller mutates its own webhook configurations (KSV-0114). This is what the operator is *for*. New CC: `operator-purpose-bound-rbac`.
- `kube-state-metrics` (both `minikube-indri` and `k3s-ringtail`) holds `list/watch` on secrets to expose `kube_secret_info` and `kube_secret_labels` metrics. KSM's metric schema only reads metadata, never the `data:` field. New CC: `kube-state-metrics-metadata-only`.
- **Over-broad RBAC → fix**
- `grafana-clusterrole` had `get/watch/list` on `secrets` because the dashboard-sidecar config used `RESOURCE=both` (ConfigMaps + Secrets). Nothing in the cluster labels Secrets with `grafana_dashboard=1`, so this was unused power. Switched both sidecar instances to `RESOURCE=configmap` and removed `secrets` from the ClusterRole.
The IaC cronjob also did not previously pass `--mutelist-file`, which is why every IaC finding reported as unmuted regardless of mutelist configuration. The new `mutelist/iac.yaml` is bundled into the existing `prowler-mutelist` ConfigMap and mounted via `items:` selector.
## Test plan
- [ ] `kubectl --context=minikube-indri kustomize argocd/manifests/prowler/` — already passes locally
- [ ] `kubectl --context=minikube-indri kustomize argocd/manifests/grafana/` — already passes locally
- [ ] Deploy from this branch via `argocd app set prowler --revision prowler-iac-mutelist && argocd app sync prowler` and same for `grafana`
- [ ] Manually trigger the IaC cronjob and verify `MUTED=True` on the 6 critical findings (`kubectl --context=minikube-indri -n prowler create job --from=cronjob/prowler-iac-scan prowler-iac-test`)
- [ ] Restart grafana pod and confirm dashboards still render (sidecar still finds them via ConfigMap watch)
- [ ] After verify, `argocd app set <app> --revision main && argocd app sync <app>` post-merge
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #340
Previously only the K8s CIS in-cluster scan was processed; the weekly
container-image and IaC Prowler scans were running on schedule but never
reviewed. Now each scan gets its own status / severity / week-over-week
delta, with top-N grouped tables (by check ID and resource) for the
high-volume image and IaC outputs.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
No version bump; build deps (jinja2, pyyaml) still loose-pinned and fine.
Known issue: deployed v1.0.3 package predates phone-hide commit; tracked
separately in Todoist by user.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The signed offset format read as "due in 5 days" rather than
"5 days overdue", causing misreads. Switch to self-explanatory
text: "5d overdue" / "due in 2d" / "due today".
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Claude Code only auto-loads CLAUDE.md. The prose shim told agents to go
read AGENTS.md, which is easy to skip. Replacing the shim with
`@AGENTS.md` inlines AGENTS.md content into the session prompt, so the
startup rules (ai-docs, blumeops-tasks, change classification) land in
context unconditionally.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Splits the nebulous gandi-operations how-to into two single-topic cards
(manage-eblu-me-dns, rotate-gandi-pat) and adds a mise task for the
recurring _acme-challenge TXT cleanup needed due to a value-comparison
bug in libdns/gandi v1.1.0 that prevents certmagic's cleanup phase from
removing presented TXT values.
The gandi reference card is updated to drop the false "different
credential from Pulumi PAT" claim — verified during the 2026-04-27
incident that Caddy and Pulumi share a single PAT.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Captures the procedure used to restore mealie's SQLite DB from a borgmatic
archive after the post-DR wipe: extract from borg, snapshot the wiped DB,
swap via a helper pod on the ReadWriteOnce PVC, fix UID 911 ownership.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Extend (not replace) home-manager's default sway keybindings via
lib.mkOptionDefault, with lib.mkForce on the custom overrides that
conflict with defaults. Add Mod+F1 cheatsheet binding (fuzzel-filterable).
Move fuzzel's border-radius/border-width out of [main] into a proper
[border] section with the expected short names.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Now that argocd's Authentik OAuth2 client is public, `argocd login --sso`
works for day-to-day use. Promote it to the default in AGENTS.md,
argocd-cli reference, and troubleshooting; keep the admin/password flow
documented as a break-glass fallback for when Authentik is unavailable.
Also drops --grpc-web from every interactive login command — confirmed
extraneous (login succeeds without it). Left in CI workflows and
`argocd cluster add` untouched; those are different contexts that I
didn't re-test.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Now that argocd's Authentik OAuth2 client is public (PKCE-only), the
client_secret plumbing is dead code:
- delete argocd-oidc-authentik ExternalSecret and drop it from kustomization
- remove AUTHENTIK_ARGOCD_CLIENT_SECRET env from authentik-worker
- remove argocd-client-secret mapping from authentik-config ExternalSecret
The argocd-client-secret field in the 1Password "Authentik (blumeops)"
item is now unreferenced and can be deleted there.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Changes argocd's Authentik OAuth2 client from confidential to public and
drops the clientSecret from argocd-cm. Public + PKCE works for both the
web UI (argocd-server backend) and the argocd CLI (`argocd login --sso`)
without a shared secret, matching OAuth 2.1 guidance.
Confidential → public was needed because the CLI can't hold a client
secret; Authentik's per-app issuer model made the alternative
("cliClientID" pattern with separate public client) awkward since it
requires a shared issuer across apps which Authentik doesn't serve.
Follow-up: deadcode AUTHENTIK_ARGOCD_CLIENT_SECRET env wiring and the
argocd-oidc-authentik ExternalSecret once verified.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds http://localhost:8085/auth/callback to the ArgoCD OAuth2 provider's
redirect_uris so `argocd login --sso` works. Loopback redirect is the
RFC 8252 pattern for native CLI apps; PKCE (already enabled) covers the
code-interception risk.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
After dispatching, poll the Forgejo API for the run matching our
head_sha and print `mise run runner-logs <N>` so the suggested monitor
command is one copy-paste away. Falls back to the bare command if the
poll times out.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Tailscale operator still defaults to privileged proxy pods with no
seccomp profile (issue #7359 open upstream). Control remains valid.
Added note about ProxyClass + device plugin remediation path.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The upstream binary expects CWD=/app (relative config.yml lookup,
lumberjack logfile at ./log/app.log). Without this, the pod crashed on
startup — the ConfigMap-mounted /app/config.yml wasn't found and zerolog
spammed "mkdir log: permission denied" as it tried to create ./log at
/ as nonroot.
Creates /app as 1777 (tmp-style) so nonroot can write logs; WorkingDir
set to /app so the default config path resolves correctly.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Built from main in run #516 after #339 merged. Follows the navidrome
kustomization convention (deployment image = local ref + :kustomized,
kustomization override = newTag only).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
## Summary
- Mirrors `github.com/0x2142/frigate-notify` at `v0.5.4` to `forge.ops.eblu.me/mirrors/frigate-notify`.
- Adds `containers/frigate-notify/default.nix` — `buildGoModule` + `dockerTools.buildLayeredImage`, following the `ntfy` pattern.
- Uses `-tags goolm` to avoid the libolm CGO dependency (matrix notifier is imported unconditionally in the upstream but we only use ntfy alerts).
- Runs as nonroot (UID 65534), exposes port 8000, bundles `cacert`/`tzdata`.
## Why
Move `ghcr.io/0x2142/frigate-notify:v0.5.4` (ringtail-deployed) under local control. Aligns with the [[indri → ringtail migration plan]] and the `default.nix` convention for ringtail-targeted containers documented in [[build-container-image]].
## Verification
- `dagger call build-nix --src=. --container-name=frigate-notify export --path=./out.tar.gz` produces a valid 20MB docker archive (10 layers) with `blumeops/frigate-notify` tag locally.
- Hashes pinned for `fetchgit` (src) and `vendorHash` (go modules).
## Follow-up (post-merge)
1. `mise run container-build-and-release frigate-notify` — release from main SHA.
2. C0 follow-up: update `argocd/manifests/frigate/kustomization.yaml` image ref to `registry.ops.eblu.me/blumeops/frigate-notify:v0.5.4-<sha>-nix`.
3. ArgoCD auto-syncs the deployment.
## Test plan
- [ ] `dagger call build-nix` succeeds from a clean checkout.
- [ ] `mise run container-build-and-release frigate-notify --dry-run` looks correct.
- [ ] After release + kustomization swap: frigate-notify pod comes up healthy on ringtail; ntfy alerts still fire on Frigate events.
Reviewed-on: #339
Swaps the k8s runner label from the local bootstrap tag (v0.20.6-9b6be09)
to the equivalent image rebuilt by CI from main. Functionally identical;
closes the bootstrap loop.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Points the k8s Forgejo runner label at the locally-bootstrapped
runner-job-image built from the Alpine container.py on this branch.
Once merged, CI will rebuild the same image from the same SHA.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Bumps the Dagger engine/CLI from v0.20.1 to v0.20.6 (mise pin, dagger.json
engineVersion, SDK regen) and rewrites the runner-job-image container as a
native Dagger pipeline on Alpine 3.23 using the shared alpine_runtime helper,
replacing the Debian-based Dockerfile. All Forgejo Actions in this repo use
actions/checkout (a JS action), so musl is not a compatibility concern.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
## Summary
- consolidate forgejo-runner how-to docs into current cards
- upgrade the k8s forgejo-runner deployment to the latest v12.8.x runner image
- switch the k8s runner from first-boot register flow to declarative server.connections config
- keep the runner image on the native Dagger build path and update the surrounding manifests/secrets
## Notes
- PR opened early for C1 review
- implementation and deployment verification will follow in subsequent commits
Reviewed-on: #338
Historical one-shot fix from the zot hardening chain — knowledge is
self-evident in containers/ntfy/default.nix and container-version-check
regex. Should have been removed at mikado finalization. Scrubbed the two
wiki-link references in add-container-version-sync-check.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Forgejo's web action routes don't support API token auth for private
repos (only session cookies or public access). Switch log fetching to
read the zstd-compressed log files directly from indri via SSH —
Forgejo stores all runner logs on disk regardless of which runner
executed the job.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
runner-logs now always authenticates with the Forgejo API token
(via --token flag, FORGEJO_TOKEN env, or 1Password) so it works on
private repos. The --repo default is auto-detected from the git
remote origin URL instead of being hardcoded.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The static asset cache block (css/js/png/etc) was missing
proxy_set_header Host, so Caddy received "forge.eblu.me" instead of
"forge.ops.eblu.me" and couldn't route the request. HTML loaded fine
because the main location / block had the header.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
All 7 ArgoCD containers had no resource limits, allowing them to consume
unlimited CPU/memory during node pressure events. This contributed to
cluster-wide probe timeout cascades on minikube-indri.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Comprehensive docs pass reflecting the new Fly proxy architecture:
- Fly proxy routes through Caddy on indri (not per-service TS Ingress)
- Direct WireGuard peering via --port=41641 pinning
- DERP relay performance lesson in Tailscale docs
- Caddy now in public traffic path
- indri tagged as flyio-target
- Removed fly-reload references
- Updated architecture diagrams and per-service setup guide
- Added changelog fragment
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Tailscale Ingress pods in k8s can't establish direct WireGuard
connections (stuck behind pod-network NAT → DERP relay → 20s latency).
Indri's host-level Tailscale CAN peer directly with Fly.
Change all nginx upstreams to route through Caddy on indri instead of
per-service Tailscale Ingress endpoints. Tag indri as flyio-target in
the Tailscale ACL so the Fly proxy can reach it.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Enable direct peer-to-peer WireGuard connections by pinning tailscaled
to port 41641 and exposing it as a UDP service. Without this, all
traffic routes through Tailscale DERP relays causing 20+ second
latency. Requires dedicated IPv4 (allocated: 168.220.82.221).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The nix-built Alloy image sets User=65534 (nobody). Even with
privileged: true, a non-root user gets no effective capabilities
(CapEff=0). Override with runAsUser: 0 so Beyla gets CAP_BPF and
CAP_SYS_ADMIN needed for eBPF instrumentation.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
NixOS defaults kernel.unprivileged_bpf_disabled=2, which blocks BPF
syscalls outside the init namespace even with CAP_BPF. Set to 1 so
privileged containers (Beyla/Alloy tracing) can create BPF maps.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Beyla (alloy-tracing) has been failing since April 13 with
"failed to set memlock rlimit: operation not permitted" because k3s
inherits the default 8MB memlock limit. Set LimitMEMLOCK=infinity on
the k3s systemd service so privileged containers can use eBPF.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The earthdistance extension (depends on cube) must be created before
restoring the teslamate database — discovered missing after 2026-04-13 DR.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Queries the Forgejo API to verify the target commit exists on the remote
before dispatching a build, preventing wasted CI runs on unpushed commits.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replace Dockerfile with container.py for native Dagger builds.
Bump devpi-server 6.19.1→6.19.3, devpi-web 5.0.1→5.0.2.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Replace per-request DNS resolution (variable-based `proxy_pass`) with static `upstream` blocks and `keepalive` connection pools
- Reuses TLS connections through the Tailscale tunnel instead of handshaking per request
- Add `mise run fly-reload` for nginx config reload without full redeploy (re-resolves upstream DNS)
## Trade-off
DNS is resolved at config load, not per-request. If Tailscale Ingress pods get new IPs (restart, reschedule), `mise run fly-reload` is needed. A Grafana alert will be added to detect this.
## Still TODO on this branch
- [ ] Grafana alert for upstream unreachable (triggers fly-reload reminder)
- [ ] Docs pass
- [ ] Deploy from branch and verify latency improvement
- [ ] Changelog fragment
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #337
Was sending Connection: upgrade on every proxied request, which is
semantically wrong for normal HTTP traffic. Use a map to conditionally
send 'upgrade' only when the client requests a WebSocket switch,
'close' otherwise.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Previous max bucket was 10s — all slower requests collapsed into +Inf,
making p50/p90/p99 unreadable during the Forgejo archive DoS.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Crawlers follow auth redirects to /user/login which is pointless for them.
Saves round-trips for both sides.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Points kustomization at v3.8.2-7a42aeb, the first image built from the
new container.py (replacing the Dockerfile).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The task was missing FLY_API_TOKEN injection, requiring manual fly auth
login. Now uses op read to fetch the deploy token from 1Password.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replaces the hand-written Dockerfile with container.py using the shared
alpine_runtime helper, which bumps the base image from Alpine 3.22 to 3.23.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The _1password-gui package doesn't export native messaging manifests
in the format the Firefox wrapper expects. The 1Password NixOS module
already handles native messaging host registration separately.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The xdg desktop entry and Librewolf user.js prefs didn't fix the
OAuth callback hang. Try stock Firefox instead as a simpler path.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The xdg desktop entry and mimeapps were already registered but
Librewolf doesn't delegate unknown URI schemes to the system
handler by default. This adds user.js prefs to complete the chain.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Facebook has been scraping forge mirror repos at ~3-4 req/s, slowing
down the Forgejo instance. Serve robots.txt directly from nginx to
disallow /mirrors/ while leaving eblume/* accessible to crawlers.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Shared Dagger helpers (src/blumeops/) affect all Dagger-built containers,
making path-based auto-triggers unreliable. All builds now go through
`mise run container-build-and-release <name>`.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Extend go_build() with buildmode and extra_env params, migrate miniflux
and forgejo-runner to use it, and bump all Alpine bases from 3.22 to 3.23.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Update kustomization image tags to the new container.py-built images
(v4.1.1-r1-2c483ce, v1.0.1-2c483ce).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replace Dockerfiles with native container.py for both transmission and
transmission-exporter. Updates base images (Alpine 3.23, Python 3.14),
pins uv to 0.11.6 instead of :latest.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
LaunchAgents now call borgmatic directly at its mise-installed path
instead of routing through `mise x`, which triggered macOS TCC
permission dialogs (e.g. "mise wants to access Documents") that hung
headless sessions and caused backup failures.
Also adds `mise install` to the ansible role so borgmatic installation
is fully managed, and pins the version in both mise.toml and the role
defaults.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Upgrade Prowler from 5.22.0 to 5.23.0
- Remove the `enumerate-images` init container workaround from `cronjob-image-scan.yaml`
- Use native `--registry` and `--image-filter` flags now that upstream fix (PR prowler-cloud/prowler#10470) is released
The init container was a workaround for prowler-cloud/prowler#10457 where `--registry` args weren't forwarded to the provider constructor. We wrote the fix, it was merged, and v5.23.0 includes it.
## Test plan
- [ ] Build new container (`mise run container-release prowler 5.23.0`)
- [ ] Update kustomization.yaml with new image tag
- [ ] Sync prowler ArgoCD app from branch
- [ ] Manually trigger image scan job and verify `--registry` works natively
- [ ] Verify CIS and IaC scan cronjobs still work
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #336
Removed Grafana from the control description — no Prowler finding
references it. Tightened scope to match actual usage (ArgoCD wildcard
RBAC mute). Added workflow-bot scoping note.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Adds automated node-level verification to `review-compliance-reports`: kubelet file perms/ownership, kubelet config args, etcd CA separation, RBAC cluster-admin bindings
- Mutes the 14 MANUAL Prowler findings via new `manual-node-checks.yaml` mutelist file
- New `node-config-automated-verification` compensating control documents the approach
- Script fails loudly (red FAIL + verdict panel) if any check deviates from expected values
## Test plan
- [x] `mise run review-compliance-reports` — all 12 node checks PASS
- [x] Injected bad expected value (perms 400 vs actual 600) — FAIL rendered correctly
- [x] Fixed colon-in-binding-name bug (kubeadm:cluster-admins) with tab-separated jsonpath
- [ ] After merge: sync prowler mutelist ConfigMap and verify next scan shows 0 MANUAL findings
## Note
Prowler coverage is minikube-indri only — ringtail/k3s is a known gap tracked separately.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #335
## Summary
- Add native Dagger `container.py` for forgejo-runner (Go + Alpine runtime, static binary with CGO for SQLite)
- Update kustomization to point to local registry image (tag is placeholder until CI builds)
- Uses existing `clone_from_forge("forgejo-runner", ...)` mirror
## Test plan
- [x] `dagger call build --src=. --container-name=forgejo-runner` passes locally
- [ ] CI container build from branch succeeds
- [ ] Update kustomization tag to built image, deploy from branch via ArgoCD `--revision`
- [ ] Verify runner registers and picks up jobs
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #334
The lockfile bakes in devpi URLs — Dagger does a locked install, not
fresh resolution. This is the mechanism behind the cold-cache failure.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
After a DR rebuild, devpi's empty cache causes race conditions under
concurrent load — metadata is served but wheel files 404. Also deploys
the first container.py-built teslamate image.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- New how-to: rebuild-minikube-cluster with full bootstrap procedure
validated during 2026-04-13 DR event
- Update restart-indri: warn about minikube delete, macOS permission
dialog on first Tailscale SSH, forgejo_actions_secrets dep cycle
- Update disaster-recovery reference: link to rebuild procedure
- Update CLAUDE.md: never run minikube delete
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Discovered during DR that paperless was the only service DB not backed
up by borgmatic. Uses same blumeops-pg cluster on port 5432.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The authentik-redis image is nix-built on ringtail (amd64 only) and was
previously running under QEMU emulation on arm64 minikube. Discovered
during DR recovery when fresh minikube lacked binfmt registration.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The Dagger engine's internal OTLP proxy returns 500 on /v1/metrics when
there's no real backend, causing ~9s retry warnings per pipeline step.
Point OTEL_EXPORTER_OTLP_ENDPOINT at Tempo to give it a real endpoint.
Also removes the stale os.environ workaround from main.py (the SDK
initializes telemetry before our module loads, so it had no effect).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The Dagger engine shim sets OTEL_METRICS_EXPORTER before our module
loads, so os.environ.setdefault was a no-op. Switch to a hard override.
Remove the redundant workflow-level env var since the fix belongs in
the module.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Upgrade grafana-sidecar from 1.28.0 to 2.6.0 (the 2.x memory regression #462 is resolved; ~35MB static overhead is acceptable)
- Port build from Dockerfile to native Dagger container.py
- Add liveness/readiness probes using the new /healthz endpoint on port 8080
- Update docs to reflect container.py migration and remove stale pin note
## Test plan
- [ ] Build container: `mise run container-build-and-release grafana-sidecar`
- [ ] Update kustomization tag with new image tag
- [ ] Deploy from branch: `argocd app set grafana --revision grafana-sidecar-2.6.0 && argocd app sync grafana`
- [ ] Verify sidecar health endpoint: `kubectl exec -n monitoring <pod> -c grafana-sc-dashboard -- wget -qO- http://localhost:8080/healthz`
- [ ] Verify dashboards load in Grafana UI
- [ ] `mise run services-check`
Reviewed-on: #332
The control claimed all images came from the private registry, but 12+
services pull from external public registries. Updated description to
reflect reality and catalogued external-image categories in notes.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Outdated leaf card removed; zot.md now links to new service-versions
reference card instead. Added reverse link from review-services.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replace broken SSH+filesystem log retrieval with Forgejo web API
endpoint. Fix CLI to use run numbers (not task IDs), add --repo
for querying any forge repo (e.g. sporks), --limit/-n for listing
size. Document runner-logs as the way to verify build success in
CLAUDE.md and container build docs.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Point miniflux kustomization at the main-built v2.2.19-138e23d image
(replacing the branch tag). Disable the OTLP metrics exporter at module
import time to prevent ~11s retry delays in CI — the env var must be set
inside the module, not the runner shell, because the SDK runs inside the
Dagger engine container.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The Dagger Python SDK's OTLP metrics exporter hits a non-functional
local endpoint (500s), burning ~9s per retry cycle. Set
OTEL_METRICS_EXPORTER=none in the build-dagger CI job.
Also update navidrome kustomization to the main-SHA tag (c86b5d7).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## 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
Add flyio-tailscale (v1.94.1), flyio-nginx (1.29.6-alpine), and
flyio-alloy (v1.14.1) entries with new `fly` service type so future
upgrades go through the service-review workflow.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Tailscale :stable pulled v1.96.5 during last deploy, which returns
SERVFAIL for tailnet DNS names (no upstream resolvers set). This broke
all public routing (forge/docs/cv.eblu.me) through the Fly proxy.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
check_alert() used head -1 to display only the first firing instance,
silently swallowing additional alerts (e.g. frigate pod-not-ready was
hidden behind ollama).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The general rate limit zone used $binary_remote_addr (Fly's internal
proxy IP), causing all external clients to share one bucket. Switch to
$http_fly_client_ip to match forge_auth's correct behavior.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The Pulumi code has had a forge.eblu.me CNAME since it was added, but the
doc's DNS table only listed docs and cv. Also fixed the __main__.py
description to mention CNAMEs alongside A records.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
preview.quality was at the top level (invalid); moved under record
with a valid preset (very_low). Also fix services-check to catch
Grafana "Alerting (NoData)" state which was silently passing.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Bug-fix release with web UI fixes, LDAP page size, and SAML SLO
redirect. Also bumps client-go to v3.2026.2.1.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Previews are ~4MB/hour at default quality (CRF 1), served over NFS from
sifaka. Reducing to CRF 8 shrinks preview files to improve review page
load times when scrubbing older footage.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Strip redundant "unifi-poller-" prefix from generated slugs, bringing
UIDs from 45-48 chars down to 32-35 chars.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
C0 follow-up to #327: update from branch-SHA tags to main-SHA tags
after squash-merge rebuild.
indri: v2.18.0-f59f885
ringtail: v2.18.0-f59f885-nix
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Build kube-state-metrics v2.18.0 locally from forge mirror, replacing upstream `registry.k8s.io` image
- Dockerfile (two-stage Go build) for indri/minikube
- default.nix (buildGoModule + buildLayeredImage) for ringtail/k3s
- Both kustomization files updated with `newName` pointing to local registry
## Verification
- [x] Nix build succeeded on ringtail (`nix-build` → 10-layer image)
- [x] Dockerfile build succeeded locally (`dagger call build` → ~2min)
- [x] `container-version-check --all-files` passes (2.18.0 consistent across Dockerfile, nix, service-versions.yaml)
- [ ] CI builds container images from this branch
- [ ] Update kustomization `newTag` with SHA-tagged version from CI
- [ ] ArgoCD sync on both clusters
## Test plan
- Trigger CI build: `mise run container-build-and-release kube-state-metrics`
- Verify tags: `mise run container-list kube-state-metrics`
- Update newTag in kustomization files with CI-produced tag
- Sync ArgoCD on indri: `argocd app sync kube-state-metrics`
- Sync ArgoCD on ringtail: `argocd app sync kube-state-metrics --context=k3s-ringtail` (note: argocd uses its own auth, not kubectl context)
- Verify metrics still flowing to Prometheus
Reviewed-on: #327
Worker forks 4 Dramatiq processes each loading the full Django app
(~250MB each), hitting the 1Gi limit on startup. Ringtail has ample
RAM headroom.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
teslamate had superuser on the shared blumeops-pg cluster (which also
hosts miniflux and authentik). Downgraded to plain database owner with
extension ownership (cube, earthdistance) transferred manually so it
can still ALTER EXTENSION UPDATE. earthdistance is untrusted in PG so
DROP+CREATE would need temporary superuser escalation.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Doc review found the authentik database, user, and external secret were
missing, along with the immich-pg borgmatic secret. Added Cluster column
to Users table for clarity. Set last-reviewed: 2026-04-07.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Patch upgrade with bug fixes (diff normalization, installation ID cache).
Pin the upstream manifest URL to commit SHA for supply chain integrity.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Rich markup parser interprets [text] as style tags, stripping
wiki-links like [[review-compensating-controls]] to empty [].
Escape description lines with rich.markup.escape().
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Verified: tailscale serve status shows only svc:k8s, ACLs restrict
tag:flyio-target to port 443 with admin/operator ownership only,
indri has no flyio-target tag. All 10 muted findings remain valid.
Noted gap: no automated alerting on new flyio-target devices.
Tracked in Todoist as MC4 (Manual Compliance Control Check CronJob).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Resolves 4 unmuted Prowler core_seccomp_profile_docker_default
findings on alloy, immich-server, immich-machine-learning, and
immich-valkey.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
New mise task fetches Prowler reports from sifaka, parses with proper
muted/unmuted distinction, shows week-over-week delta, and includes
a scaffold for Kingfisher once JSON/CSV output is available upstream.
Moved all legacy top-level reports on sifaka into date subdirectories
to match the current CronJob output structure. Updated
read-compliance-reports doc with task reference and links.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The `--exclude` flag added in #321 never existed in nix — it was
introduced broken and never tested. Replace with dynamic input
discovery: query `nix flake metadata --json` for all input names,
filter out skip_inputs (default: nixpkgs-services), pass the rest
as positional args. Also bump NIX_IMAGE 2.33.3 → 2.34.4.
Updated inputs: nixpkgs, home-manager, disko.
nixpkgs-services stays pinned.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
All Helm chart mirrors (grafana-helm-charts, connect-helm-charts,
cloudnative-pg-charts) have been deleted from forge.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The storage-provisioner is a bare Pod with no controller. If the node
restarts via Docker Desktop (rather than `minikube start`), kubelet
restores static pods but bare pods are lost. Detect this and re-run
`minikube start` to restore addons.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replace the Helm chart deployment with plain kustomize manifests following
the Authentik pattern (separate deployments per component). Consolidate
the immich-storage ArgoCD app into the main immich app. Add no-helm-policy
doc establishing kustomize as the standard deployment mechanism.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
PR #10470 merged 2026-03-30; initContainer workaround stays until a
Prowler release includes the fix (latest is 5.22.0).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Gamescope 3.16.17 segfaults on NVIDIA 580.x in nested Wayland/Sway due
to explicit sync issues (ValveSoftware/gamescope#1662). Use a Sway
window rule to force RDR2 fullscreen instead.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Move Steam config from configuration.nix to a dedicated gaming.nix module
and add gamescope for fullscreen/resolution management with Proton games.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Add `containers/tempo/Dockerfile` — two-stage Go build from forge mirror, modeled on loki
- Switch kustomization from upstream `grafana/tempo` to `registry.ops.eblu.me/blumeops/tempo`
- Bump Tempo 2.10.1 → 2.10.3
## Test plan
- [ ] Kick off container build via `mise run container-build-and-release tempo`
- [ ] Update kustomization `newTag` with built image tag
- [ ] Deploy from branch: `argocd app set tempo --revision local-tempo-container && argocd app sync tempo`
- [ ] Verify Tempo health: `curl tempo.ops.eblu.me/ready`
- [ ] Verify traces flowing in Grafana Tempo datasource
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #323
Stamp single-user-cluster last-reviewed to 2026-04-01 after verifying
Tailscale ACLs and kubeconfig distribution. Add aspirational how-to card
documenting what PCI DSS evidence collection would look like (CCW,
artifacts, Drata workflow). Link from existing review process card.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Add `nixpkgs-services` flake input pinned to a specific nixpkgs commit, with an overlay that pulls `forgejo-runner`, `snowflake`, and `k3s` from it instead of the rolling `nixpkgs`
- Dagger `flake-update` pipeline now excludes `nixpkgs-services` via `--exclude`
- Fix stale nix-container-builder version in service-versions.yaml (was 12.6.4, actually running 12.7.2)
- Add k3s and minikube to service-versions.yaml tracking
- Document the pinning approach in review-services how-to and ringtail reference
## Motivation
During service review, discovered that flake updates had silently upgraded forgejo-runner from 12.6.4 → 12.7.2 without updating service-versions.yaml. This "sneak-in upgrade" bypasses the service review process. The overlay ensures these three services only change versions deliberately.
## Test plan
- [ ] Verify `nix flake update` from `nixos/ringtail/` does not change `nixpkgs-services` lock entry
- [ ] Verify `mise run provision-ringtail` builds successfully with the overlay
- [ ] Confirm running service versions unchanged after deploy
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #321
Patch upgrade picks up idempotent FetchTask API, offline registration
fix, cloudflare/circl security dep update, and custom gRPC user-agent.
No config defaults changed.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Restrict backup to library/ and upload/ only (skip regenerable encoded-video/,
thumbs/, backups/). Add SSH ServerAliveInterval to prevent broken pipe on long
transfers, and checkpoint_interval so interrupted backups save progress.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Kingfisher exits 200 (findings) or 205 (validated findings) on success.
Normalize these to 0 so the CronJob completes instead of restarting.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Mirror repos cause scan failures (likely ephemeral storage or timeout).
Scan only eblume/ repos until we investigate the root cause.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Container runs as user 65534 (nobody) but /tmp was owned by root.
Set sticky bit + world-writable (1777) like a standard /tmp.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Kingfisher needs a writable temp directory for git clones and scanning.
Nix containers don't create /tmp by default.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Nix containers don't include a shell by default. The CronJob needs
/bin/bash for the inline script that generates timestamped filenames.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Kingfisher will build via Nix on ringtail instead of Dockerfile on
indri, so the skip is no longer needed.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Kingfisher's Rust + Boost/vectorscan build exhausts indri's memory
(aws-sdk-ec2 alone needs 2-3GB for rustc). Build locally on Gilbert
and push manually until we have a beefier build host.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The pod-level RuntimeDefault seccomp profile (07e9c81) overrides the
DinD sidecar's privileged flag in newer Kubernetes versions, blocking
Docker daemon syscalls. Set Unconfined explicitly on the DinD container
while keeping RuntimeDefault on the runner container.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Bail if upstream already has branches named 'blumeops' or 'deploy',
which would conflict with the spork branch naming strategy.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Upstream can push workflows (in .github/ or .forgejo/) that execute
on our runners via any trigger mechanism including cron. Runner label
mismatch is the current defense but is fragile. No complete fix exists
short of disabling Actions entirely.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
git checkout <branch> is ambiguous when both origin and mirror remotes
have the same branch name. Use -B to explicitly create from origin.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Forks from mirror repos have has_actions disabled by default.
PATCH the repo settings to enable it so the mirror-sync workflow runs.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Check local path, mirror existence, and fork absence upfront.
Fail fast with clear error messages before touching forge or disk.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Trying to add remotes to an existing clone gets the origin wrong.
Better to error out and let the user handle it.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Spork-create mise task sets up a floating-branch soft-fork of a
mirrored upstream project with daily mirror-sync via Forgejo Actions.
Includes explanation card, how-to guides for setup and branch
management, and the spork-create uv script.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Deploys MongoDB Kingfisher as a weekly CronJob on minikube-indri
- Scans all Forgejo repos (eblume + all orgs) for leaked secrets with live validation
- Produces timestamped HTML and JSON reports on sifaka NFS (`/volume1/reports/kingfisher/`)
- Forgejo API token sourced from 1Password via ExternalSecret
- Uses official `ghcr.io/mongodb/kingfisher:1.91.0` container image
- Runs Sunday 4am (after Prowler's 3am k8s scan)
## Resources
- CronJob, PV/PVC (sifaka NFS), ExternalSecret
- ArgoCD Application with manual sync + CreateNamespace
## Test plan
- [x] Sync ArgoCD `apps` app to pick up new kingfisher Application
- [x] Set `--revision feature/kingfisher-cronjob` on kingfisher app
- [x] Verify ExternalSecret creates the `kingfisher-forgejo-token` Secret
- [x] Trigger manual job: `kubectl create job --from=cronjob/kingfisher kingfisher-manual -n kingfisher --context=minikube-indri`
- [ ] Verify reports appear on sifaka at `/volume1/reports/kingfisher/`
- [ ] After merge: set `--revision main` and re-sync
Reviewed-on: #317
Running alongside TruffleHog to compare coverage. Kingfisher uses
staged-only mode with validation disabled for fast, offline-safe
pre-commit checks. Validation will be enabled in the planned cron job.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Sifaka's Tailscale can revert to userspace networking after package
updates, causing NFS mounts to fail because the NFS daemon sees
127.0.0.1 instead of the client's Tailscale IP. Added troubleshooting
how-to doc and updated sifaka reference card with frigate export and
TUN requirement.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Resources were under wrong Helm value keys (server.resources,
machine-learning.resources) and never applied to pods. Move to correct
bjw-s chart paths (*.controllers.main.containers.main.resources).
Increase liveness/readiness probe timeouts from 1s to 5s to prevent
kubelet from killing healthy-but-busy pods during ML inference load.
Remove CPU limits (keep requests only) to avoid throttling.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Adds a second borgmatic config (`photos.yaml`) that backs up `/Volumes/photos` (sifaka SMB mount, ~128 GB) to a dedicated BorgBase repo (`immich-photos`), running daily at 4 AM
- Separate launchd agent (`mcquack.eblume.borgmatic-photos`) so photo backups run independently from the main backup
- Refactors `borgmatic_metrics` script to support multiple repos with a `repo` Prometheus label
- Updates Grafana "Borg Backups" dashboard with a `repo` template variable so you can filter/compare repos
- Docs updated: `backups.md`, `borgmatic.md`
## Prerequisites (manual)
- [x] Create `immich-photos` repo on BorgBase with same SSH key
- [ ] Upgrade BorgBase plan to Small ($24/yr) if currently on free tier (128 GB exceeds 10 GB limit)
- [ ] After deploy: `borg init` the new repo (borgmatic does this automatically on first run)
## Test plan
- [ ] Dry run: `mise run provision-indri -- --check --diff --tags borgmatic,borgmatic_metrics`
- [ ] Deploy borgmatic role and verify both configs deployed
- [ ] Run `borgmatic --config ~/.config/borgmatic/photos.yaml create --verbosity 1` manually for first backup (will take hours)
- [ ] Verify metrics script collects from both repos: `~/.local/bin/borgmatic-metrics && cat /opt/homebrew/var/node_exporter/textfile/borgmatic.prom`
- [ ] Sync grafana-config in ArgoCD and verify dashboard repo selector works
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #315
## Summary
- Add `authentik` database (blumeops-pg cluster) to borgmatic pg_dump backups
- Add `immich` database (immich-pg cluster) to borgmatic pg_dump backups
- For immich-pg: new borgmatic managed role with `pg_read_all_data`, ExternalSecret, Tailscale LoadBalancer service, and Caddy L4 TCP proxy on port 5433
- Update backup docs to reflect all four CNPG databases + mealie SQLite
## Deploy plan
Deploy order matters — k8s resources must exist before ansible can route to them:
1. **ArgoCD (databases app):** sync to pick up immich-pg borgmatic role, ExternalSecret, and Tailscale service
```
argocd app set blumeops-pg --revision feature/borgmatic-all-pg-backups
argocd app sync blumeops-pg
```
2. **Wait** for `immich-pg-tailscale` service to get a Tailscale IP and `immich-pg.tail8d86e.ts.net` to resolve
3. **Ansible (caddy):** deploy Caddy L4 route for port 5433
```
mise run provision-indri -- --tags caddy
```
4. **Ansible (borgmatic):** deploy updated config and .pgpass
```
mise run provision-indri -- --tags borgmatic
```
5. **Verify:** trigger a manual borgmatic run and check all four pg_dump streams succeed
```
borgmatic --verbosity 1 2>&1 | grep -E '(Dumping|ERROR)'
```
## Test plan
- [x] `kubectl kustomize` builds cleanly
- [x] `ansible --check --diff` for borgmatic and caddy show expected changes
- [ ] ArgoCD sync succeeds for databases app
- [ ] `immich-pg.tail8d86e.ts.net` resolves
- [ ] `pg.ops.eblu.me:5433` accepts connections
- [ ] `borgmatic --verbosity 1` dumps all four databases without errors
Reviewed-on: #314
Single-file Go tool implementing the QArt technique (Russ Cox, 2012)
using only the public rsc.io/qr API. Generates QR codes whose data
modules form a recognizable image by exploiting error correction
freedom via GF(2) Gaussian elimination.
Includes a web UI with live-updating sliders for version, mask,
rotation, dx/dy offset, and scale. Keyboard shortcuts for rapid
iteration. Also works as a CLI for batch generation.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Update manage-lockfile doc with post-deploy steps (kernel update detection,
reboot guidance, generation pruning). Add prune-ringtail-generations mise
task that keeps the 5 most recent generations plus the most recent one
matching the booted kernel for safe rollback.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Fix stale CV service doc (URL, forge domain, container tag) and add
guidance for reviewing build-time dependencies in private forge repos
during service reviews.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The 5-minute lookback window kept stale data from terminated pods
visible during rollouts, causing the alert to sit in Pending for
~5 minutes after every routine deployment. 60s still covers two
scrape cycles (30s interval) while clearing stale data much faster.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Reduced `for` from 30m to 5m and lookback window from 5m to 1m. The old
values caused alerts to linger long after apps returned to Synced state.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
branch-cleanup: Add PROTECTED_PREFIXES with preserve/* exclusion so
preserved work-in-progress branches are never deleted.
observability.md: Document Pyroscope profiling work on branch
preserve/pyroscope-profiling/pr-313, blocked on ringtail kernel
sysctl settings (kptr_restrict=0, perf_event_paranoid≤1). Also
document Faro/RUM as future potential with privacy considerations.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Minor release with new widgets (Tracearr, SparklyFitness), Seerr rename,
and dependency bumps. No breaking changes for our config.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Upgrade External Secrets Operator from v1.3.2 (helm-chart-2.0.0) to v2.2.0
- Migrate from Helm chart deployment to static kustomize manifests, matching the repo's kustomize-first pattern
- Merge separate `-config` ArgoCD apps into the main operator apps (6 → 4 apps)
- Clean up Helm-specific labels (`helm.sh/chart`, `managed-by: Helm`)
- Update README example from v1beta1 to v1 API
## Breaking changes assessment
Low risk — v2.0.0 removed Alibaba and Device42 providers (we use neither). No templating changes affect us. All ExternalSecrets already use v1 API.
## Deployment steps
1. Sync CRDs first on both clusters (new CRD version)
2. Sync operator apps (now kustomize-based)
3. Verify ClusterSecretStore and all ExternalSecrets are healthy
4. Delete orphaned config apps: `argocd app delete external-secrets-config` and `-config-ringtail`
5. `mise run services-check`
Reviewed-on: #312
## Summary
- Add Snowflake proxy as a native systemd service on ringtail (NixOS)
- Uses `pkgs.snowflake` from nixpkgs (v2.11.0)
- Hardened systemd unit with DynamicUser, ProtectSystem=strict, 512MB memory limit
- Prometheus metrics enabled on localhost:9999
## What is Snowflake?
A Tor pluggable transport that helps censored users reach the Tor network via WebRTC. **This is NOT a Tor exit node** — traffic exits through Tor exit nodes operated by others. The proxy operator cannot see traffic content (double-encrypted) and destination servers never see the proxy's IP.
## Changes
- `nixos/ringtail/configuration.nix` — new systemd service definition
- `docs/reference/services/snowflake-proxy.md` — service reference card
- `docs/reference/infrastructure/ringtail.md` — updated systemd services section
- `service-versions.yaml` — added entry (type: nixos)
## Deploy plan
After review, deploy via `mise run provision-ringtail`. Service starts automatically.
## Test plan
- [ ] `mise run provision-ringtail` succeeds
- [ ] `ssh ringtail 'systemctl status snowflake-proxy'` shows active
- [ ] `ssh ringtail 'journalctl -u snowflake-proxy --no-pager -n 20'` shows broker connections
- [ ] `ssh ringtail 'curl -s localhost:9999/metrics'` returns Prometheus metrics
Reviewed-on: #311
The kubectl image lacks curl/python3. Use the prowler image
(which has Python) with a pure-Python urllib script instead.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Prowler's --registry flag doesn't work (registry args not passed
to ImageProvider constructor, prowler-cloud/prowler PR #10128
regression). Use an init container to enumerate images from the
zot catalog API and generate an image list file instead.
See: https://github.com/eblume/prowler/tree/fix/image-provider-registry-args
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Image scan: add https:// scheme to registry URL.
IaC scan: use --scan-repository-url (Prowler clones the repo
itself), removing the need for an init container. The flag
is --scan-path for local dirs, --scan-repository-url for git.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Clone repo in init container, scan Dockerfiles and K8s manifests
with Prowler's IaC provider (Trivy). Reports written to
sifaka:/volume1/reports/prowler-iac/.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add Trivy to the Prowler container for image and IaC scanning.
New CronJob (Saturday 3am) scans all blumeops/* images in the
registry for CVEs, embedded secrets, and Dockerfile misconfigs.
Reports written to sifaka:/volume1/reports/prowler-images/.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
C0 follow-up to #309: update kustomization newTag for all containers
rebuilt by the merge (authentik, authentik-redis, ntfy, alloy).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Replace upstream `docker.io/library/redis:7-alpine` (Redis 7.4.8) with a nix-built container using Redis 8.2.3 from nixpkgs
- Introduce **attached service pattern**: `parent` field in service-versions.yaml, `<parent>-<component>` naming convention, and `assert pkgs.redis.version == version` in default.nix to prevent silent version drift on `flake.lock` updates
- Document the pattern in [[review-services]] so future attached services slot in cleanly
- Backfill `parent: grafana` on existing `grafana-sidecar` entry
## Version drift protection
1. `flake.lock` update bumps nixpkgs redis → `assert` in `default.nix` breaks `nix-build`
2. Developer updates `version` in `default.nix` → prek's `container-version-check` demands matching `service-versions.yaml` update
3. Both must agree before commit succeeds
## Test plan
- [ ] Build container from branch on ringtail (`mise run container-build-and-release authentik-redis`)
- [ ] Update kustomization `newTag` to branch-built image tag
- [ ] Sync authentik ArgoCD app from branch (`argocd app set authentik --revision localize-redis && argocd app sync authentik`)
- [ ] Verify Authentik login, session persistence, and task queue still work
- [ ] After merge: C0 follow-up to update `newTag` to the main-built image tag
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #309
Bump from RC to latest stable (security fixes for config endpoint and
cross-camera auth). Add new 0.17 motion retention tier at 365 days,
reduce continuous from 180 to 30 days.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Uses the grafana_folder annotation to place the dashboard in the
existing folder created by alert rule provisioning.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Two panels: currently firing alerts (firing/pending/noData/error) and
recent state changes. Refreshes every 30s. Uses Grafana's built-in
alertlist panel type — no datasource needed.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Dramatiq defaults to one worker process per CPU core. On ringtail (16 cores)
this spawned 16 processes, each loading the full Django app, exceeding the
1Gi memory limit and causing a crash loop (228 restarts over 7 days).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Merges `build-container.yaml` and `build-container-nix.yaml` into a single workflow
- Detect job classifies each changed container by presence of `Dockerfile` and/or `default.nix`
- Dockerfile containers build on `k8s` (indri) via Dagger; Nix containers build on `nix-container-builder` (ringtail) via nix-build + skopeo
- Containers with both build files (alloy, nettest, ntfy) get built on both runners
## Test plan
- [ ] Push a change to a Dockerfile-only container (e.g. grafana) — verify it builds on k8s only
- [ ] Push a change to a nix-only container (e.g. jobsync) — verify it builds on nix-container-builder only
- [ ] Push a change to a dual container (e.g. ntfy) — verify it builds on both runners
- [ ] Test workflow_dispatch with a specific container name
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #306
Replace hardcoded image tags in Quick Reference tables with pointers to
kustomization manifests (tags drift with every container release). Fix
Prometheus CNPG scrape target, remove misleading .ts.net URLs, expand
external-secrets stub, add backup/disaster-recovery cross-references.
Limit doc-reviewer agent to one doc per cycle.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Remove `group: ""` from ignoreDifferences in tailscale-operator and
tailscale-operator-ringtail — ArgoCD normalizes away the empty string
field, so the live state never matches git.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The Fly container pulls from tailscale/tailscale:stable which is still
v1.94.2. The `tailscale wait` command doesn't exist until v1.96.2.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
v1.96.3 exists as a GitHub release but Docker Hub images for both
tailscale/tailscale and tailscale/k8s-operator haven't been published
yet (v1.94.2 is still latest). Revert the image tags; the fly/start.sh
`tailscale wait` improvement and review date stamps are retained.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Bump Tailscale operator, proxy containers, and init containers from v1.94.2 to v1.96.3 across both clusters (indri + ringtail via shared base kustomization)
- Replace hand-rolled `until tailscale status` polling loop in `fly/start.sh` with `tailscale wait --timeout 60s` (new in v1.96.2)
- Stamp kube-state-metrics review date (already current at v2.18.0)
## Notable upstream changes (v1.94.2 → v1.96.3)
- Go upgraded from 1.25 to 1.26
- `tailscale wait` command — blocks until daemon is running + interface has IP
- AuthKey policy now applies only when users are not logged in (behavioral change)
- Peer Relay improvements (metrics, EC2 IMDS, UDP socket scaling)
- UPnP stability fixes
## Deploy plan
1. Merge PR
2. Sync tailscale-operator on indri: `argocd app sync tailscale-operator`
3. Sync tailscale-operator on ringtail: `argocd app sync tailscale-operator-ringtail --server ringtail...`
4. Verify proxy pods roll with new image: `kubectl --context=minikube-indri -n tailscale get pods`
5. Verify ingress connectivity (spot-check a few `*.tail8d86e.ts.net` services)
6. Rebuild + deploy Fly proxy container (separate step, picks up `tailscale wait` change)
## Test plan
- [ ] ArgoCD diff looks clean for both apps before sync
- [ ] Proxy pods on indri come up healthy with v1.96.3 images
- [ ] Proxy pods on ringtail come up healthy with v1.96.3 images
- [ ] Tailscale ingress services remain reachable (e.g., grafana, prometheus)
- [ ] Fly proxy rebuild deploys successfully with `tailscale wait`
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #304
## Summary
Mikado chain to replace `mise run services-check` with Grafana Unified Alerting backed by ntfy push notifications.
**Design:**
- Grafana Unified Alerting evaluates rules against Prometheus/Loki
- ntfy webhook contact point delivers iOS notifications
- Anti-noise policy: page once per 24h per alert group
- Every alert links to a runbook in `docs/how-to/alerts/`
- services-check eventually queries the alerting API instead of doing its own probes
**Chain (bottom-up):**
1. `configure-grafana-alerting-pipeline` — enable alerting, ntfy contact point, notification policy
2. `first-alert-and-runbook` — end-to-end proof of concept with blackbox probe failure
3. `port-services-check-alerts` — migrate all services-check probes to alert rules + runbooks
4. `refactor-services-check-to-query-alerts` — rewrite services-check to query Grafana API
5. `deploy-infra-alerting` — goal card
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #303
Replace single aggregate camera_fps check with per-camera FPS validation
and NFS storage accessibility check. Motivated by an outage where Frigate
API responded OK but NFS mount was inaccessible, causing "no frames" in UI.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Also catches kiwix's transmission sidecar up from v4.0.6-r4 to
v4.1.1-r1, matching the torrent service (upgraded in PR #282 but
the kiwix sidecar was missed). No breaking changes — old RPC
protocol is supported through 4.x.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Point all services at the 613f05d images which carry the new
consistent OCI labels. Skipped kiwix/transmission (old v4.0.6-r4
version, no matching build) and docs/quartz (no 613f05d build).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Every container now carries title, description, version, source, and
vendor labels per the OCI image spec. Version is derived from the
existing CONTAINER_APP_VERSION ARG at build time.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The top-level "uid" in Grafana dashboard JSON is at 2-space indent
near the end of the file, not the first occurrence. Match on ^ "uid"
to avoid clobbering nested datasource uid references.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The previous sed replaced ALL "uid" fields in dashboard JSON files,
including datasource references inside panels, causing dashboards to
go dark. Scope the replacement to only the first occurrence (the
top-level dashboard UID) using GNU sed 0,/pattern/ addressing.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add init container to pre-populate ConfigMap dashboards before Grafana
starts, eliminating the race between the sidecar and the provisioner
that caused dashboard DB records to be deleted and re-created with new
IDs. Also stamp stable UIDs on TeslaMate and UnPoller dashboards
fetched from upstream.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Four project-scoped subagents that formalize existing mise task
workflows as constrained, specialized AI agents:
- infra-health: background health monitor (wraps services-check)
- doc-reviewer: persistent-memory documentation reviewer
- change-classifier: C0/C1/C2 triage before work begins
- mikado-navigator: C2 chain state advisor (wraps docs-mikado)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
StatefulSet volumeClaimTemplates are immutable and minikube's hostpath
provisioner doesn't enforce PVC size limits anyway. Add comments noting
the data grows freely on the 1.8TB backing disk.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Rename dashboard title since borgmatic is just the execution layer.
Add Backup Duration Over Time panel next to New Data Per Backup.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Forgejo: show only notifications and pull requests
- Jellyfin: show only movies/series/episodes, hide now playing
- Grafana: hide data sources, show dashboards and alerts only
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add useEqualHeights: true so service tiles within each row expand to
match the tallest tile, fixing uneven layout from widget metrics.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
style: row makes each group span the full page width (one per row),
while columns: 4 tiles services horizontally within each group.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add maxGroupColumns: 1 so each category gets its own full-width row,
with service tiles arranged side-by-side within each group.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Move NVR, Jellyfin, and DJ to new Home group. Move Grafana from Content
to Infrastructure. Switch all layout groups from column to row style.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The Mealie SQLite dump hook used `minikube-indri` (the context name on
gilbert), but on indri itself the context is just `minikube`. This caused
the before_backup hook to fail, aborting all backups since the hook was added.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Increase retention: continuous 3→180d, detections 14→30d, alerts 30→730d.
Plenty of NFS headroom (~9.4 TiB free, ~6.6 GB/day for one camera).
Add frigate-recording check to services-check that verifies camera_fps > 0,
which would have caught the 6-day outage from the mqtt config removal.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Frigate's config schema requires an `mqtt` field even when MQTT isn't
used. Commit 40f1568 removed it along with Mosquitto, causing Frigate
to fail validation on startup. Add `mqtt.enabled: false` to satisfy
the schema without needing a broker.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Point alloy-k8s at v1.14.0-61f02a0 (Dockerfile) and both ringtail
deployments at v1.14.0-61f02a0-nix (Nix build).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Add `containers/alloy/` with dual Dockerfile + Nix build files for Grafana Alloy v1.14.0
- Both builds fetch source from forge mirror (`forge.ops.eblu.me/mirrors/alloy.git`), build the web UI (Node), then compile the Go binary with `netgo embedalloyui` tags
- Update all three alloy deployments (alloy-k8s, alloy-ringtail, alloy-tracing-ringtail) to use `registry.ops.eblu.me/blumeops/alloy`
- `promtail_journal_enabled` tag omitted — requires systemd headers and none of our configs use `loki.source.journal`
## Build verification
- **Dockerfile:** Tested locally via `docker build`, binary reports `v1.14.0` with correct tags
- **Nix:** Tested on ringtail via `nix-build`, all three hashes (fetchgit, npmDeps, goModules) resolved and build succeeds
## Post-merge steps
1. Wait for CI to build the container from main (both Dockerfile and Nix workflows)
2. `mise run container-list alloy` to find the `[main]` tagged image
3. C0 follow-up to update `newTag` in all three kustomizations from `v1.14.0-placeholder` to the real tag
4. Sync ArgoCD apps and verify pods come up healthy
Reviewed-on: #300
Mealie's orderBy=random requires a paginationSeed parameter, otherwise
the API returns 422. Added the seed to all random query examples.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Agent-facing guide for generating unified cooking timelines from
Mealie meal plans. Covers querying the API, picking balanced meals
(protein/carb/vegetable), and interleaving recipe steps into a
relative timeline so everything finishes together.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Enable recipe parsing from images/photos, ingredient extraction, and
URL scraping via OpenAI API (gpt-4o). Rename ExternalSecret from
mealie-oidc to mealie-secrets to hold both OIDC and OpenAI credentials.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Deploy UnPoller as a k8s service on indri to export UniFi controller metrics to Prometheus
- Custom-built container from forge mirror (`containers/unpoller/Dockerfile`)
- Credentials pulled from 1Password via external-secrets
- Prometheus scrape job added, docs and service-versions updated
## Test plan
- [ ] Build container: `mise run container-release unpoller v2.34.0`
- [ ] Update kustomization tag with built image tag
- [ ] Deploy from branch: `argocd app set unpoller --revision feature/unpoller && argocd app sync unpoller`
- [ ] Verify pod connects to UX7 controller (check logs)
- [ ] Confirm `unpoller` target appears in Prometheus
- [ ] Query `unifi_` metrics in Grafana
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #298
## Summary
- Upgraded borgmatic from 2.0.13 to 2.1.3 on indri (via mise/pipx)
- Key changes: improved borg warning handling, memory/performance improvements, `source_directories_must_exist` now defaults to true (already set in our config)
- Verified: config validates, dry-run passed against both sifaka (local) and borgbase (offsite) repos
## Borg Warnings Investigation
The main concern was 2.1.0's change to treat borg warnings as errors. In 2.1.3 this was partially reverted — "file not found" warnings (exit code 107) are back to being warnings. Our config already sets `source_directories_must_exist: true`, and all four source directories were verified present on indri.
## Test plan
- [x] `borgmatic --version` confirms 2.1.3
- [x] `borgmatic config validate` passes
- [x] `borgmatic create --dry-run` succeeds against both repositories
- [x] All source directories verified present on indri
- [ ] Verify next scheduled backup (2:00 AM) completes successfully
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: #297
Mark run-1password-backup and troubleshooting as reviewed. Troubleshooting
gets inline wiki-links for all referenced services, a new ringtail/k3s
section, and a cross-reference to restart-indri.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Consolidate 4 Authentik Nix derivation docs into one card
(authentik-nix-build-components.md)
- Merge build-grafana-container + build-grafana-sidecar into
build-grafana-images.md
- Move agent-change-process from how-to/ to explanation/ (it's a
methodology doc, not a task guide)
- Extract Caddy custom build section from reference card into
how-to/deployment/build-caddy-with-plugins.md
- Move expose-service-publicly from how-to/ to tutorials/ (it's a
comprehensive walkthrough, not a quick task reference)
- Update all wiki-link references across affected docs
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- ArgoCD ref: correct Git Source URL to forge.ops.eblu.me:2222
- Authentik ref: add Zot as active OIDC client, blueprint, and secret
- Federated login: remove Zot from Future Work (completed in PR #236)
- devpi/start.sh: use bash array for command building (proper quoting)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
ai-sources now skips docs/ to avoid duplicating ai-docs output.
CLAUDE.md notes ai-sources as available for deep context on problems
with a large surface area.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
ai-sources: concatenates all git-tracked files (excluding lock files
and other non-useful artifacts) for full codebase AI context (~353K
tokens).
ai-docs: now concatenates all docs instead of a curated subset (~85K
tokens).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Replaces 18 TeslaMate dashboard ConfigMaps (713 KB / 22,080 lines) with a Grafana init container
- Init container fetches dashboard JSON directly from `mirrors/teslamate` on forge, pinned to `v3.0.0`
- Grafana's file provider picks them up from `/tmp/dashboards/TeslaMate/` via `foldersFromFilesStructure`
- Non-TeslaMate dashboards remain as ConfigMaps (unchanged)
## How it works
- New `init-teslamate-dashboards` init container uses busybox `wget` to fetch each JSON file from `https://forge.eblu.me/mirrors/teslamate/raw/tag/v3.0.0/grafana/dashboards/`
- Files land in `/tmp/dashboards/TeslaMate/`, same emptyDir volume the sidecar uses
- The sidecar continues to handle ConfigMap-based dashboards; the init container handles TeslaMate
- Version pin is in the init container args (TESLAMATE_VERSION)
## Deployment and Testing
- [ ] Sync `grafana` app from branch — verify init container runs and fetches dashboards
- [ ] Sync `grafana-config` app from branch — verify TeslaMate ConfigMaps are pruned
- [ ] Check Grafana UI: TeslaMate folder should still contain all 18 dashboards
- [ ] Verify non-TeslaMate dashboards are unaffected
- [ ] After merge: sync both apps from main
Reviewed-on: #296
Caddy v2.11 (#7454) auto-rewrites the Host header to match the
upstream address for HTTPS backends. This causes services behind
Tailscale Ingress to see *.tail8d86e.ts.net instead of *.ops.eblu.me,
breaking Authentik OAuth flows, Homepage host validation, and other
services that check the Host header.
Only apply header_up for HTTPS backends (Tailscale Ingress); HTTP
backends (forge, registry, jellyfin, sifaka) are unaffected.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Mirrors `tailscale/tailscale` on forge (`mirrors/tailscale`)
- Replaces vendored `operator.yaml` (495 KB / 5,386 lines) with ArgoCD apps sourcing the upstream static manifest, pinned via `targetRevision: v1.94.2`
- Adds `tailscale-operator-base` app for indri and `tailscale-operator-base-ringtail` for ringtail
- Local kustomization retains only ProxyClass and DNSConfig custom resources
- Updates `[[tailscale-operator]]` doc to reflect new sourcing
## Deployment and Testing
- [ ] Register `mirrors/tailscale` repo in ArgoCD (it needs to know about the new repo)
- [ ] Sync `apps` app to pick up the new `tailscale-operator-base` app definitions
- [ ] Sync `tailscale-operator-base` — verify CRDs, RBAC, operator Deployment come up
- [ ] Sync `tailscale-operator` — verify ProxyClass, DNSConfig still apply cleanly
- [ ] Verify existing Tailscale Ingresses still work (ProxyGroup pods healthy)
- [ ] Repeat for ringtail cluster
- [ ] After merge: apps already point at tags, no revision reset needed
Reviewed-on: #295
- Caddy is now a mcquack LaunchAgent, not brew services
- Add missing Jellyfin and Caddy to shutdown commands and autostart list
- docs-preview: accept paths with or without docs/ prefix
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
- Upgrade zot OCI registry from v2.1.13 to v2.1.15 on indri
- Addresses CVE-2025-30204 (golang-jwt memory) and open redirect via callback_ui
- No config template changes needed (externalUrl is auto-allowlisted)
- Requires Go 1.25.7 (bump from 1.25.6 via mise)
## Data Safety
- Data directory ~/erichblume/zot is NOT touched during build or deploy
- No schema migrations in v2.1.14 or v2.1.15
- Storage format remains OCI spec 1.1.0
## Deployment Steps
- [ ] SSH to indri: bump Go to 1.25.7 via `mise use go@1.25.7`
- [ ] Fetch and checkout v2.1.15 in ~/code/3rd/zot
- [ ] Build: `mise x -- make binary`
- [ ] Restart LaunchAgent
- [ ] Verify: `curl -s http://localhost:5050/v2/` returns 200
- [ ] Verify: `curl -s https://registry.ops.eblu.me/v2/_catalog` lists repos
- [ ] Verify: `mise run services-check`
Reviewed-on: #293
The 27B Q4_K_M model needs ~7.3 GiB system RAM for CPU-offloaded layers
but only 6.8 GiB was available within the 22Gi cgroup. Bumping to 24Gi
and enabling flash attention (reduces KV cache memory) should provide
enough headroom.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The 27B Q4_K_M model is ~17 GB, exceeding the 16 GB VRAM on the RTX 4080
by ~1 GB. Ollama will offload a few layers to CPU RAM, so the pod memory
limit needs headroom beyond the previous 16Gi.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Mosquitto has been dormant since frigate-notify switched from MQTT to
webapi polling (529ba10). Tear down live infra (ArgoCD app, namespace)
and remove all manifests, service-versions entry, services-check, and
doc references.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
1Password adds account ID and timestamp to export filenames. The script
now globs ~/Documents for .1pux files instead of expecting a fixed name.
Also fixes a Rich markup error with bracket characters in the prompt.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Review build-jobsync-container.md: fix nonexistent `mirror-sync` task
reference (Forgejo mirrors sync automatically), mark reviewed
- Remove bat hint from docs-review checklist (output not visible in
agent sessions), keep docs-preview hint as user-facing step
- Simplify review-documentation.md visual preview section
- Fix Python 3.14 tarfile deprecation warning in docs-preview
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
New `mise run docs-preview <card>` task builds docs via Dagger and serves
them locally in the production quartz container (image parsed from ArgoCD
kustomization), opening the browser directly to the specified card.
Container auto-cleans after 1 hour.
Also updates docs-review checklist and review-documentation how-to to
reference the visual preview workflow.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Add jobsync pod check (ringtail k3s) and HTTP endpoint to `services-check`
- Add JobSync entry to homepage dashboard under new "Apps" group
- Mark jobsync as reviewed at v1.1.4 (current with upstream)
- Changelog fragment added
## Deployment and Testing
- [ ] Sync homepage app from branch: `argocd app set homepage --revision review/jobsync && argocd app sync homepage`
- [ ] Verify JobSync appears on go.ops.eblu.me dashboard
- [ ] Run `mise run services-check` to verify new checks pass
- [ ] After merge: `argocd app set homepage --revision main && argocd app sync homepage`
Reviewed-on: #291
## Summary
Fixes the Facebook crawler spider trap that's been generating infinite recursive URLs like `/how-to/tutorials/tutorials/how-to/explanation/...` for several days.
**Root cause:** Quartz SPA mode + nginx `try_files` fallback to `index.html` meant any fabricated URL returned the root HTML shell with HTTP 200. Crawlers followed relative links from those fake URLs, creating infinite recursion.
**Fix:**
- Disable Quartz SPA mode (`enableSPA: false`) — all pages are now fully static HTML
- Replace nginx SPA fallback with `=404` + Quartz's static `404.html`
- Remove `robots.txt` exclusions (no longer needed)
**Docs cleanup (Obsidian.nvim compat no longer needed):**
- Delete hand-curated category index files (`tutorials.md`, `reference.md`, `how-to.md`, `explanation.md`) — Quartz auto-generates folder pages
- Delete `postgresql-storage.md` (redirect stub) and `migrate-forgejo-from-brew.md` (stale history)
- Drop `docs-check-index` and `docs-check-filenames` prek hooks
- Rewrite `docs-check-links` to allow path-based wiki-links (`[[path/to/file]]`) and only error on true ambiguity
- Add `ai-docs` doc tree listing to replace index files for AI context
- Add natural cross-links from reference cards to fix orphan docs
## Deployment and Testing
- [ ] Merge and let the build pipeline run
- [ ] Verify docs.eblu.me serves pages correctly with full page loads
- [ ] Verify non-existent URLs return 404
- [ ] Monitor crawler traffic — should drop to near zero for fabricated URLs
Reviewed-on: #290
The previous robots.txt had a typo blocking /explorer/ instead of
/explore/, allowing Facebook's crawler to hit the spider trap.
Also block /tags/ which has the same infinite relative-link issue.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Add JobSync service reference card (`docs/reference/services/jobsync.md`) with architecture, secrets, observability, and JSearch API docs
- Add JobSync and Ollama to ringtail's workloads table (both were missing)
- Add JobSync to the reference index
- Wire `RAPIDAPI_KEY` through ExternalSecret and deployment env var for JSearch job search automation
- Document Loki log queries for observability (no metrics endpoint exists)
- Update deploy-jobsync how-to with new env var, observability section, and reference card link
## Deployment and Testing
- [ ] Sign up for RapidAPI JSearch API (free tier: 500 req/month)
- [ ] Add `rapidapi_key` field to "JobSync" 1Password item
- [ ] Merge PR
- [ ] `argocd app sync jobsync` to pick up new env var
- [ ] Verify job search works at https://jobsync.ops.eblu.me/dashboard/automations
Reviewed-on: #289
## Summary
C2 Mikado chain to deploy [JobSync](https://github.com/Gsync/jobsync) — a self-hosted job application tracker — to ringtail's k3s cluster.
### Mikado Graph
```
deploy-jobsync (goal)
├── build-jobsync-container
│ └── mirror-jobsync
└── integrate-jobsync-ollama
```
### What is JobSync?
Next.js app with SQLite for tracking job applications. Features resume management, application pipeline tracking, and AI-powered resume review/job matching.
### Key Decisions
- **Ringtail k3s** (not minikube-indri) — colocates with Ollama for zero-latency AI
- **Nix container** via `buildLayeredImage` — no Dockerfile, mirrors upstream source on forge
- **Ollama for AI** — uses existing deployment, no API keys needed for AI features
- **No upstream fork** — vanilla JobSync, Anthropic AI deferred to future work if needed
### Current Status
Planning phase — cards committed, ready for review before implementation begins.
Reviewed-on: #288
A close commit with zero preceding impl commits is valid — some leaf
nodes involve operational steps (e.g., creating a mirror) with no code
changes. Removed the false-positive check.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- 1Password Connect uses non-standard numeric log levels (`1`=error, `2`=warn, `3`=info, `4`=debug, `5`=trace) per [1Password/connect#44](https://github.com/1Password/connect/issues/44)
- Alloy extracts the `level` JSON field as-is, so info-level health checks get `level="3"` in Loki
- Grafana expects string level labels — numeric values are unrecognized, causing misclassified log severity/coloring
- Adds a `stage.match` + `stage.template` in the Alloy pipeline scoped to `{namespace="1password"}` to normalize numeric levels to standard strings
- Other services are completely unaffected (scoped by namespace, not global)
## Deployment and Testing
- [ ] Sync alloy-k8s from branch: `argocd app set alloy-k8s --revision fix/onepassword-numeric-log-levels && argocd app sync alloy-k8s`
- [ ] Wait ~2 minutes for new logs to flow
- [ ] Verify level labels: `curl -sG "http://localhost:3100/loki/api/v1/label/level/values" --data-urlencode 'query={namespace="1password"}'` should show `"info"` and `"warn"` instead of `"3"` and `"2"`
- [ ] Check Grafana log panel for 1password namespace — logs should no longer appear as errors
- [ ] After merge: `argocd app set alloy-k8s --revision main && argocd app sync alloy-k8s`
Reviewed-on: #287
- Add last-reviewed date
- Align service type sections with actual types (argocd/ansible/nixos)
- Remove nonexistent "Helm Chart" and "Hybrid" sections
- Fold custom container guidance into ArgoCD section
- Reference kustomization.yaml for image tags instead of Helm charts
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add last-reviewed date
- Replace raw pulumi commands with mise task equivalents
- Reference C0/C1/C2 change classification for making changes
- Note that prek handles link validation automatically
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Frigate 0.17 does not auto-create clips/previews/<camera>/, causing
review page previews to silently fail with 500 errors.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Phase 2 of Dagger upgrade: bump engine version, update runner
deployment to v0.20.1-24f7512, and fix docs reference card version.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Phase 1 of Dagger upgrade: update the CLI in the runner container first
so CI can build the new image with the old engine version. See
[[upgrade-dagger]] for the full procedure.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Documents the correct two-phase upgrade procedure to avoid the
chicken-and-egg problem where CI can't build its own replacement.
Also fixes outdated version references in the Dagger reference card.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The /explorer/ SPA endpoints were the source of all spider-trap traffic.
A robots.txt Disallow is a better fix than serving 404s — it prevents
crawlers from entering the infinite URL tree in the first place, avoids
serving large numbers of 404s that hurt SEO, and doesn't break legitimate
deep links.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Bare image references in manifests were ambiguous — unclear whether the
tag was intentionally omitted or managed by kustomize. Add :kustomized
sentinel to all 37 image refs overridden by kustomize images transformer.
Add sync notes for tailscale-operator proxyclass (CRD fields not processed
by kustomize). Mark devpi reviewed (6.19.1 is current).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Removed descriptions, table formatting, and Mikado chain commentary
from the how-to index — it should be links only. Added last-reviewed
date.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Built locally to break the chicken-and-egg: the old runner couldn't
build its own replacement because it needed Dagger 0.20.0.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The Dagger module was upgraded to v0.20.0 in d15071a but the runner job
image still had the old CLI, causing build-blumeops to fail with a
version mismatch.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The external-secrets operator adds conversionStrategy, decodingStrategy,
and metadataPolicy defaults to the live object, causing perpetual
OutOfSync in ArgoCD. Declare them explicitly to match.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Add Authentik OAuth2 provider + application blueprint for ArgoCD (ringtail side)
- Add OIDC config to ArgoCD ConfigMap with Authentik as identity provider (indri side)
- Map Authentik `admins` group to ArgoCD `role:admin` via RBAC policy
- ExternalSecrets on both sides pull `argocd-client-secret` from 1Password
- Local admin password remains as break-glass — both login methods coexist
## Pre-deployment manual step
Add `argocd-client-secret` field to "Authentik (blumeops)" in 1Password with a random value (e.g., `openssl rand -hex 32`).
## Deployment order
1. Sync Authentik app on ringtail first (blueprint + secret + worker env var)
2. Sync ArgoCD app on indri second (cm, rbac, ExternalSecret)
## Verification
- [ ] `argocd-client-secret` field added to 1Password
- [ ] Authentik app synced on ringtail — blueprint applied, provider created
- [ ] ArgoCD app synced on indri — OIDC config applied
- [ ] SSO login works: visit `https://argocd.ops.eblu.me` → "Log in via Authentik" → admin access
- [ ] Break-glass: local admin/password login still works
Reviewed-on: #284
Minor upstream release with doc and CI fixes. Also corrects kiwix.md
to reference the actual custom registry image and torrents.txt path.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Dashboard "Download/Upload Rate by Torrent" panels were querying
transmission_torrent_download_bytes (total_size * percent_done) and
transmission_torrent_upload_bytes (uploaded_ever) — cumulative byte
gauges, not rates. Added new metrics using Transmission's native
rate_download/rate_upload and updated dashboard queries.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The count-only stat wasn't actionable. New table shows pod name, container,
restart count, and memory limit for each OOMKilled container. Waiting reason
panel narrowed to make room.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
OOMKilled containers previously only appeared briefly in "Unhealthy Pods"
while dying, then vanished on restart. New panels use persistent metrics
(last_terminated_reason) and restart rate tracking.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
During finalization, all mikado frontmatter (requires, status, branch) should
be removed — cards become plain documentation linked via wiki-links. Updated
agent-change-process docs and cleaned up 10 cards from closed chains. Also
fixed ai-docs referencing deleted plans/ files.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The plans/ directory predated the mikado method approach. Deleted all
completed and abandoned plans, converted the still-relevant
migrate-forgejo-from-brew into a lean mikado chain root card under
how-to/forgejo/, cleaned up dangling wiki-links across docs, and
fixed a stale "pre-commit" reference to "prek".
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Upgrade Transmission from 4.0.6-r4 to 4.1.1-r1
- Uses Alpine edge community repo for transmission packages, keeping stable alpine:3.22 base
- Fix stale image reference in service doc (was linuxserver, now custom registry image)
- Mark transmission as reviewed in service-versions.yaml
## Context
Service review found Transmission two minor versions behind (4.0.6 → 4.1.1). Alpine 3.22 only packages 4.0.6, so transmission is installed from edge's community repo with an exact version pin.
4.1.0 added improved µTP performance, IPv6/dual-stack UDP tracker, JSON-RPC 2.0 API. 4.1.1 is a bugfix release (20+ fixes).
Dagger test build passed locally.
## Deployment and Testing
- [ ] Build container via Forgejo workflow (`mise run container-build-and-release transmission`)
- [ ] Update kustomization.yaml with new image tag
- [ ] `argocd app set torrent --revision feature/transmission-review && argocd app sync torrent`
- [ ] Verify web UI at https://torrent.ops.eblu.me
- [ ] Check Grafana Transmission dashboard still receives metrics
- [ ] After merge: `argocd app set torrent --revision main && argocd app sync torrent`
## Note
The transmission-exporter sidecar (OOMKilling every ~30min, 294 restarts) is being tracked separately as a future replacement project.
Reviewed-on: #282
C0 changes have no branch name, so `main.<type>.md` fragments collide.
Switch to towncrier's `+<slug>.<type>.md` orphan convention and rename
existing `main.*` fragments.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The database was at /config/frigate.db (emptyDir, ephemeral) instead of
/db/frigate.db (PVC, persistent). Every pod restart wiped the database,
losing all recording history and leaving orphaned files on NFS.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
ONNX detector + CUDA ffmpeg + workers consume ~1.9Gi at steady state,
causing intermittent OOMKills at the 2Gi limit.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Add two-stage Dockerfile for Loki (Go build → Alpine runtime) in `containers/loki/`
- Rewrite kustomize image to `registry.ops.eblu.me/blumeops/loki`
- Tag is `v3.6.5-placeholder` until first CI build; will be updated post-build
## Details
- UID 10001 matches existing StatefulSet `securityContext` (runAsUser/fsGroup)
- CGO_ENABLED=0, ldflags embed version via `github.com/grafana/loki/v3/pkg/util/build`
- Clones from `forge.ops.eblu.me/mirrors/loki` (mirror created this session)
- Pattern follows miniflux (two-stage Go) + prometheus (ldflags)
## Deployment and Testing
- [ ] Trigger container build: `mise run container-build-and-release loki`
- [ ] Update kustomize tag to actual build tag
- [ ] Deploy from branch: `argocd app set loki --revision feature/loki-container && argocd app sync loki`
- [ ] Verify `/ready` endpoint and log ingestion
- [ ] After merge: update to `[main]` tag (C0 follow-up)
Reviewed-on: #280
The bash parameter expansion `${var/pat/rep}` treats `\/` in the
replacement as a literal backslash-slash, not an escaped delimiter.
This produced URLs like `https:\/\/eblume:...` instead of
`https://eblume:...`, breaking Forgejo's URL parser (500 on mirror
settings pages) and preventing mirror syncs.
Use prefix stripping (`${var#prefix}`) instead.
All 22 corrupted mirrors have been repaired on indri.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The NixOS firewall was blocking pod-to-host TCP traffic because only
tailscale0 was trusted. Pods could ping the host but not reach the
API server (port 6443), breaking Tailscale Ingress TLS cert refresh
and all ringtail services (authentik, frigate, ntfy, ollama).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Misfiled fragment from feature/ branch created a subdirectory under
changelog.d/ which towncrier doesn't support. Move the fragment to the
correct flat location and add a changelog-check mise task + prek hook
to prevent this from happening again.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Was the only app still using https://forge.eblu.me (public proxy) for
git polling. All other apps already use the internal SSH endpoint at
forge.ops.eblu.me.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The k8s and ringtail runners were hitting forge.eblu.me (fly.io proxy)
for every FetchTask poll (~every 2s), round-tripping through the public
internet unnecessarily. Use forge.ops.eblu.me (Caddy on indri, tailnet)
for infrastructure workloads.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The doc listed a nonexistent configmap.yaml instead of the actual raw
config files (grafana.ini, datasources.yaml, provider.yaml) consumed
by kustomization.yaml's configMapGenerator. Added last-reviewed date.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Prevents OOM when switching between models — only one 14B model
fits in 16GB VRAM at a time with KV cache for context.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Replace pre-commit with [prek](https://github.com/j178/prek), a faster Rust-native drop-in alternative
- Migrate config from `.pre-commit-config.yaml` (YAML) to `prek.toml` (TOML)
- Add new built-in checks: case conflicts, private key detection, executable shebangs
- Install prek via mise native registry (`aqua:j178/prek`) instead of pipx
- Update all doc references across README, contributing guide, and how-to docs
## Notes
- `check-yaml` still uses the remote `pre-commit-hooks` repo because prek's builtin fast path doesn't support `--unsafe` yet (needed for Ansible custom YAML tags)
- All existing custom hooks (docs validation, container version check, mikado invariant, workflow validation) work unchanged
- Tested: all hooks pass on clean tree, deliberate doc link breakage is caught
## Test plan
- [x] `prek run --all-files` passes all checks
- [x] Broken wiki-link correctly caught by `docs-check-links`
- [x] taplo-format auto-fixes TOML formatting on commit
- [x] commit-msg hook (mikado invariant) fires correctly
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/276
The mikado-branch-invariant-check hook now inspects staged files (commit-msg
hook) and historical commit files (standalone mode) to reject impl commits
that touch markdown files with Mikado frontmatter (requires:, status:, or
branch: mikado/). Cards should only be modified by plan, close, or finalize.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Fix go-server-derivation: wrong path target (webui not authentik-django)
and missing internal/web/static.go patch. Remove stale DRF fork content
from mirror-build-deps (no longer needed as of 2026.2.0). Add
last-reviewed to all 5 cards without it.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Python 3.14's EntryPoints uses string keys, not integer indices.
eps[0] raises KeyError(0); use next(iter(eps)) instead.
Verified on ringtail with the actual venv python.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Django's BASE_DIR is $out but source lives in site-packages. Code
like BASE_DIR / "authentik" / "sources" / "scim" / "schemas" / ...
needs a top-level symlink to find data files alongside Python source.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
lifecycle/ak uses ${TMPDIR}/authentik-mode — without TMPDIR set it
tries to write /authentik-mode in root, which user 65534 can't do.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
buildLayeredImage doesn't create /tmp by default. The container runs
as user 65534 (nobody) which can't mkdir /tmp at runtime.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Update image tag to v2026.2.0-efa9806-nix — the first source-built
authentik container from the build-authentik-from-source chain.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Create goal card and 4 prerequisite cards for building authentik from a
custom Nix derivation instead of using pkgs.authentik from nixpkgs. This
removes the dependency on the nixpkgs packaging timeline and gives full
version control over authentik releases.
Chain: mikado/authentik-source-build
Leaf nodes: authentik-api-client-generation, authentik-python-backend-derivation
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace floating :18 tag with pinned :18.3 (upstream out-of-cycle
release fixing 18.2 regressions). Stamps service as reviewed.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The LLM should read the file itself using its tools rather than
receiving it inline in the task output.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The external-secrets webhook injects conversionStrategy, decodingStrategy,
and metadataPolicy defaults on admission. Declaring them explicitly prevents
ArgoCD SSA from flagging the resource as OutOfSync.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Add `cluster` label (indri/ringtail) to all Prometheus scrape jobs, Alloy k8s metrics/logs, and Alloy host metrics/logs
- Deploy kube-state-metrics on ringtail's k3s cluster (ArgoCD app + manifests)
- Deploy Alloy on ringtail to collect pod metrics and logs, remote-writing to indri's Prometheus and Loki
- Replace single-cluster "Minikube Kubernetes" and "K8s Services Health" dashboards with:
- **Kubernetes Clusters** dashboard — multi-cluster with `cluster` and `namespace` template variables
- **Ringtail (k3s)** dashboard — dedicated ringtail view with GPU usage panels
## Deployment and Testing
1. Sync `apps` on indri ArgoCD to pick up new app definitions (`kube-state-metrics-ringtail`, `alloy-ringtail`)
2. Sync `prometheus` → verify `cluster` label on scraped metrics
3. Sync `alloy-k8s` → verify `cluster=indri` on remote-written metrics and logs
4. Run `mise run provision-indri -- --tags alloy` → verify `cluster=indri` on host Alloy metrics/logs
5. Sync `kube-state-metrics-ringtail` → verify pods running on ringtail
6. Sync `alloy-ringtail` → verify pods running, check Prometheus for `kube_pod_info{cluster="ringtail"}`
7. Sync `grafana-config` → verify dashboards appear, cluster variable populates both values
8. Check Loki for `{cluster="ringtail"}` logs from ringtail pods
## Notes
- Alloy on ringtail uses `insecure_skip_verify=true` for TLS to Prometheus/Loki (Tailscale-managed certs not in container trust store) — tighten later
- DNS resolution for `*.tail8d86e.ts.net` from ringtail pods depends on CoreDNS inheriting host's MagicDNS resolver; may need CoreDNS forwarding rules if pods can't resolve
- The old services dashboard (blackbox probes) is removed — those probes are still running in alloy-k8s and the data is still in Prometheus, just not in a dedicated dashboard
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/270
Parked car was being re-detected every few minutes at night due to IR
illumination noise triggering motion detection. Restrict the driveway
zone to [person, dog, cat] so cars and birds no longer create events
there. Cars still alert via the driveway_entrance zone.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- **mirror-create**: Auto-includes GitHub PAT from 1Password for authenticated upstream fetches at mirror creation time
- **mirror-update-pats**: New mise task that SSHes into indri and rewrites the git remote URL in every GitHub mirror's bare repo config to embed the PAT. Idempotent, supports `--dry-run`
- **app.ini.j2**: Explicit `[mirror]` section with `DEFAULT_INTERVAL = 8h` and `MIN_INTERVAL = 10m` (bakes in the defaults for visibility)
- **manage-forgejo-mirrors**: New how-to doc covering mirror creation, PAT storage, the `mirror-update-pats` task, and the full 20-day PAT rotation procedure
## Context
GitHub tightened unauthenticated rate limits for git clone/fetch in May 2025. With 23 GitHub mirrors syncing every 8 hours, authenticated fetches avoid throttling. The PAT is stored in 1Password (`Forgejo Secrets` → `github-mirror-pat`) and has been applied to all existing mirrors.
## Deployment and Testing
- [x] `mirror-update-pats` dry-run verified (23 mirrors detected)
- [x] `mirror-update-pats` applied to all 23 GitHub mirrors on indri
- [x] Idempotency confirmed (re-run shows 0 updated, 23 skipped)
- [ ] Provision indri with `--tags forgejo` to apply `[mirror]` config
- [ ] Trigger a manual mirror sync and verify success in Forgejo UI
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/269
The --style=header --color=never --decorations=always flags are now built
into the script so callers can just run `mise run ai-docs`. Also adds a
note to CLAUDE.md to never truncate the output.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Point ArgoCD app directly at forge-mirrored upstream repo (`mirrors/cloudnative-pg`) instead of the Helm charts repo
- Use `directory.include` to select the specific release manifest (`cnpg-1.27.1.yaml`) from the `releases/` directory
- No vendored files, no Helm — upgrades are a two-line change (`targetRevision` + `directory.include`)
- Delete unused `values.yaml` (was empty, all Helm defaults)
## Deployment and Testing
- [ ] Register mirror repo in ArgoCD: `argocd repo add ssh://forgejo@forge.ops.eblu.me:2222/mirrors/cloudnative-pg.git --ssh-private-key-path <key>`
- [ ] `argocd app set cloudnative-pg --revision feature/cnpg-direct-source && argocd app sync cloudnative-pg`
- [ ] Verify operator pod running: `kubectl get pods -n cnpg-system --context=minikube-indri`
- [ ] Verify CRDs exist: `kubectl get crd --context=minikube-indri | grep cnpg`
- [ ] Verify existing clusters healthy: `kubectl get clusters -A --context=minikube-indri`
- [ ] After merge: `argocd app set cloudnative-pg --revision main && argocd app sync cloudnative-pg`
## Notes
- The forge mirror was created via `mise run mirror-create` from `https://github.com/cloudnative-pg/cloudnative-pg.git`
- ArgoCD may need the mirror repo added to its known repositories if the credential template doesn't already match `mirrors/*`
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/268
The panel queried frigate_camera_events but the actual metric exposed
by Frigate is frigate_camera_events_total with a "camera" label
(not "camera_name").
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Widen `repo-creds-forge` URL prefix from `/eblume/` to host-wide `/` so it matches repos in all forge orgs (fixes `mirrors/` repos not getting SSH credentials)
- Update 8 ArgoCD app definitions from `eblume/<mirror>` → `mirrors/<mirror>` (immich-charts, cloudnative-pg-charts, external-secrets, connect-helm-charts)
- Fix stale alloy clone comment in Ansible defaults
- Bump immich v2.5.2 → v2.5.6 (bug-fix patches only)
- Update ArgoCD README bootstrap command and credential docs
## Context
Mirrors were migrated from `forge.ops.eblu.me/eblume/` to `forge.ops.eblu.me/mirrors/` in commit `cd57814`. Container Dockerfiles and image tags were updated, but ArgoCD app definitions and the repo credential template were missed, causing `ComparisonError` on apps that source Helm charts from mirrored repos.
## Deployment
1. Sync the ArgoCD `argocd` app first (picks up the widened credential template)
2. Sync the `apps` app (picks up new repo URLs for all 8 apps)
3. Verify immich resolves its ComparisonError: `argocd app get immich`
4. Sync immich to deploy v2.5.6: `argocd app sync immich`
5. Spot-check: `argocd app get external-secrets`, `argocd app get cloudnative-pg`, `argocd app get 1password-connect`
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/266
Categorized reference of all mise tasks with descriptions. Added to
the tools section of the reference index and to the ai-docs context
priming script.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Creates mirrors in the mirrors/ Forgejo org via API. Supports
GitHub, Codeberg, and generic git URLs with auto-detection.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Prometheus and teslamate builds still in progress — will update in a
follow-up commit once their 33b7f0f tags land.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Created miniflux mirror at mirrors/miniflux. All three containers
now clone from forge.ops.eblu.me/mirrors/ instead of GitHub directly.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
AirPlay from Main to IoT VLAN (Samsung Frame TV) required adding
established/related, AirPlay port, and dynamic reverse port rules —
but the root cause was rule ordering (allows appended after blocks).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The hand-written configmap.yaml had app.kubernetes.io/name and
app.kubernetes.io/instance labels; configMapGenerator dropped them.
Add options.labels to both generator entries to restore parity.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Move hardcoded image tags to kustomization.yaml `images:` transformer across **22 services** — image names in manifests become version-agnostic templates, with tags centralized in one place per service
- Replace hand-written ConfigMap manifests with `configMapGenerator:` in **12 services** — config data extracted to standalone files, generated ConfigMaps include content hashes that trigger automatic pod rollouts on changes
- Create new `kustomization.yaml` for **forgejo-runner** and **nvidia-device-plugin** (switches ArgoCD from directory mode to kustomize mode, rendered output identical)
### Services modified
**Images only (8):** cv, devpi, docs, kube-state-metrics, miniflux, navidrome, teslamate, torrent
**Images + configMapGenerator (10):** alloy-k8s, forgejo-runner, frigate, grafana, homepage, kiwix, loki, mosquitto, ntfy, prometheus
**Images only, no configMapGenerator (4):** authentik (skip blueprints — special YAML tags), tailscale-operator-base (Deployment only, CRD image fields left as-is)
**Skipped entirely (6):** argocd (remote upstream), databases (no image fields), external-secrets, grafana-config (cross-kustomization dashboards), immich (Helm-managed), 1password-connect/cloudnative-pg (no kustomization.yaml)
### What changes at deploy time
- **images:** — no functional diff, `kustomize build` produces identical output with tags
- **configMapGenerator:** — ConfigMap names gain hash suffixes (e.g., `prometheus-config` → `prometheus-config-6f42fhctcb`) and all Deployment/StatefulSet/DaemonSet references are updated automatically. Pods will restart once per service on first sync due to the name change
## Test plan
- [x] `kubectl kustomize` builds all 30 service directories successfully
- [x] Image tags verified in rendered output for all modified services
- [x] ConfigMap hash suffixes verified in rendered output
- [x] ConfigMap references in Deployments/StatefulSets confirmed to use hashed names
- [x] All pre-commit hooks pass (yamllint, shellcheck, prettier, etc.)
- [ ] `argocd app diff` each service to confirm only expected ConfigMap name changes
- [ ] Deploy from branch starting with a low-risk service (e.g., mosquitto)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/264
Grafana 12.x's grafana-postgresql-datasource plugin requires the
database name in jsonData, not just the top-level database field.
Without it, the frontend blocks all queries with "no default database
configured", causing all TeslaMate panels to show "No Data."
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The INI parser was stripping outer single quotes from
role_attribute_path = 'Admin', causing Grafana to evaluate 'Admin'
as a JMESPath field identifier instead of a string literal. This
resulted in all OAuth users getting the default Viewer role.
Replaced with a proper group-based expression that checks for the
'admins' Authentik group and maps to Admin/Viewer accordingly.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Prometheus: v3.9.1-74029e1 [branch] -> v3.9.1-2ba5d8a [main]
Grafana: v12.3.3-09ac36b [branch] -> v12.3.3-d05d2fb [main]
These images were built during PR development and referenced branch
commits that won't survive branch cleanup. The [main] tags are
identical rebuilds from the squash-merge commit.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
After investigating deployed container images, confirmed that squash-merging PRs orphans the commit SHAs embedded in container image tags. Two of our currently deployed images (prometheus, grafana) reference branch commits not on main.
This PR:
- Documents the squash-merge SHA orphan problem and the post-merge workflow in [[build-container-image]]
- Adds step 9 to the C1 process: after merging a PR that changes `containers/`, do a follow-up C0 to point manifests at the rebuilt `[main]` tag
- Rewrites `container-list` as a `uv run --script` (typer + rich + httpx)
- Adds optional container name filter (`mise run container-list prometheus` shows 10 tags instead of 4)
- Annotates every tag with `[main]` or `[branch]` based on git commit ancestry
## Test plan
- [x] `mise run container-list` — all containers shown with `[main]`/`[branch]` hints
- [x] `mise run container-list prometheus` — filtered view, more tags, correctly shows `[main]` and `[branch]`
- [x] `mise run container-list nonexistent` — error message with exit code 1
- [x] Pre-commit hooks pass
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/263
## Summary
- **End-of-cycle prompting:** After closing a leaf node and pushing, the agent should prompt the user to review and suggest ending the session rather than rushing into the next leaf
- **Reset rigor:** Reinforced that errors during impl should trigger a branch reset + plan update (not fix-forward). Documented the `git log --oneline --not main` → `git reset --hard` → `git cherry-pick` pattern with clear threshold guidance
- **`--resume` shows PR number:** Queries the Forgejo API for open PRs matching the branch, displays number/title/URL and a hint to run `pr-comments`
- **`--resume` checks git stash:** Shows stash entries as a non-presumptive hint — informs without assuming they apply
## Test plan
- [ ] `mise run docs-mikado --resume` runs without errors (no active chains case)
- [ ] On a mikado branch with an open PR, verify PR info is shown
- [ ] With stashed work, verify stash entries are displayed
- [ ] Review agent-change-process.md for clarity
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/261
The forgejo-runner container is the CI job execution environment (Dagger,
ArgoCD CLI, etc.), not the runner daemon itself. Rename to runner-job-image
to fix the version-check false positive (Dagger 0.19.11 vs daemon 12.7.0)
and clarify the distinction.
RUNNER_LABELS still references the old image name — will update after
building the image under the new name.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
actions/checkout treats short SHAs as branch name patterns, causing
fetch failures. Always resolve --ref to a full 40-char SHA before
dispatching the workflow.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
When manually dispatching a container build with --ref, the build job
now checks out the specified commit instead of the branch HEAD. This
allows building containers from feature branches before merging.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The commit-msg hook had `pass_filenames: false`, which prevented
pre-commit from passing the commit message file path. Without it,
the hook only validated existing branch history and never checked
the incoming commit against ordering rules — plan-after-impl was
silently accepted.
Remove `pass_filenames: false` so the pending commit is included
in invariant validation.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- **C0 (Quick Fix):** Now explicitly allows direct-to-main commits with no PR required — for low-risk, fix-forward-safe changes
- **C1 (Human Review):** New docs-first workflow with branch deployment (ArgoCD `--revision`, Ansible from checkout). Includes upgrade criteria for escalation to C2
- **C2 (Mikado Chain):** Introduces the **Mikado Branch Invariant** — strict commit ordering where card-introducing commits come first, followed by code progress, followed by card closures. Branch resets required when new prerequisites are discovered
Updates CLAUDE.md rules (3, 4, 8, 9) to reflect that C0 bypasses branching/PR requirements. Also updates ai-assistance-guide, how-to index, and docs-mikado task description.
## Files changed
- `CLAUDE.md` — rules and classification table
- `docs/how-to/agent-change-process.md` — full process rewrite
- `docs/tutorials/ai-assistance-guide.md` — branching and pitfalls sections
- `docs/how-to/how-to.md` — index description
- `mise-tasks/docs-mikado` — task description
- `docs/changelog.d/formalize-change-classification.doc.md` — changelog fragment
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/259
## Summary
- Delete the old 3-phase Helm chart upgrade plan (predates Mikado system)
- Create C2 Mikado chain with goal card `upgrade-grafana` and two leaf prereqs:
- `kustomize-grafana-deployment` — convert Helm to kustomize manifests
- `build-grafana-container` — home-built Grafana 12.x image (no upstream containers)
- Record first-ever Grafana review: currently at v11.4.0 on Helm chart 8.8.2
- Update service-versions.yaml, how-to index, and plans index
## Service Review Findings
- Grafana is healthy and synced in ArgoCD
- Running v11.4.0, latest upstream is 12.3.3
- Breaking changes for 12.x are low-risk (React panels only, UIDs compliant)
- PVC is disposable — dashboards and datasources are all config-provisioned
## Deployment and Testing
- [ ] No deployment needed — documentation-only change
- [ ] `docs-check-links` passes
- [ ] `docs-check-index` passes
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/258
## Summary
- Rewrites deploy-authentik from a historical changelog into a reproducible process guide
- Removes stale version info (`v1.1.2-nix`) and future work section (Forgejo federation is done, rest belongs elsewhere)
- Marks deploy-authentik as completed in plans index and completed archive
- Removes hardcoded image tag from authentik reference card (use `service-versions.yaml`)
- Adds `last-reviewed: 2026-02-23` frontmatter
## Test plan
- [x] All pre-commit hooks pass (docs-check-links, docs-check-index, etc.)
- [x] ArgoCD app verified synced and healthy
- [x] All wiki-links validated
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/257
## Summary
- Add `--progress=plain` to all `dagger call` invocations in mise tasks to prevent SIGTTOU hangs
## Root cause
Mise runs task scripts in a child process group that is not the terminal's foreground group. When `dagger call` detects a TTY (inherited from the interactive shell), it tries to render its TUI progress display, which requires terminal ioctls. Since the process is not in the foreground group, the kernel sends SIGTTOU, stopping the process indefinitely.
This only manifests when running from an interactive terminal (e.g. `pre-commit run --all-files` in fish/wezterm). CI and piped contexts are unaffected since there's no TTY.
## Changes
- `mise-tasks/validate-workflows` — add `--progress=plain`
- `mise-tasks/frigate-export-model` — add `--progress=plain`
- `mise-tasks/provision-ringtail` — add `--progress=plain`
## Test plan
- [x] `pre-commit run --all-files` completes without hanging
- [ ] Verify in interactive fish/wezterm terminal
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/256
## Summary
- `foldersFromFilesStructure` was `false` in Grafana's sidecar provider config, causing Grafana to ignore the subdirectory structure the sidecar creates from `grafana_folder` annotations
- All 18 TeslaMate dashboards were appearing in the root "Dashboards" folder despite having `grafana_folder: "TeslaMate"` annotations on their ConfigMaps
- Flipping to `true` makes Grafana replicate the sidecar's directory structure as UI folders
## Deployment and Testing
- [ ] Sync `grafana` app: `argocd app sync grafana`
- [ ] Verify TeslaMate dashboards appear under a "TeslaMate" folder in Grafana's dashboard list
- [ ] Verify other dashboards remain in the root "Dashboards" folder
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/253
## Summary
Completes the `upgrade-k8s-runner` mikado chain. Both prerequisites (workflow validation in Dagger, config review against v12 defaults) were resolved in #250.
- Bump runner image `code.forgejo.org/forgejo/runner:6.3.1` → `12.7.0`
- Update `service-versions.yaml` to track new version
- Mark goal card complete (remove `status: active`)
## Deployment and Testing
After merge:
1. `argocd app sync forgejo-runner`
2. Verify runner registers in Forgejo admin → runners
3. Trigger a test workflow (e.g. `branch-cleanup.yaml` manual dispatch)
Rollback: revert image tag to `6.3.1`, push, sync.
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/251
## Summary
- Review runner config against v12.7.0 defaults — added `shutdown_timeout: 3h`, no breaking changes found
- Add `validate_workflows` Dagger function using `forgejo-runner validate --directory .` inside upstream container
- All 6 workflows pass v12.7.0 schema validation
- Wire `mise run validate-workflows` task and pre-commit hook on `.forgejo/workflows/` changes
- Mark both leaf Mikado cards (`review-runner-config-v12`, `validate-workflows-against-v12`) complete
## Mikado State
After merge, `upgrade-k8s-runner` goal card has no unmet dependencies — ready to execute the actual image bump in a follow-up PR.
## Test Plan
- [x] `dagger call validate-workflows --src=.` passes (all 6 workflows OK)
- [x] Pre-commit hooks pass
- [ ] Reviewer: confirm `shutdown_timeout: 3h` addition to ConfigMap looks reasonable
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/250
## Summary
- C2 Mikado chain for upgrading the k8s forgejo-runner daemon (6 major versions behind)
- Root goal card with two leaf prerequisites: workflow validation and config review
- Ringtail runner is already at ~v12.6.4 via nixpkgs, no work needed there
## Mikado Chain
```
upgrade-k8s-runner (goal)
├── validate-workflows-against-v12 (leaf)
└── review-runner-config-v12 (leaf)
```
Both leaves are actionable now. The biggest risk is workflow schema validation
(introduced in v8/v9) rejecting our existing workflows.
## Next Steps
Work the leaf nodes in a follow-up session, then attempt the goal.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/249
## Summary
- Forgejo rewrites `head.ref` to `refs/pull/N/head` once a PR's source branch is deleted from the remote
- The original branch name is preserved in `head.label`
- This was causing 188 out of 246 merged PRs to go undetected by the cleanup script
- Fix: fall back to `head.label` when `head.ref` starts with `refs/pull/`
## Test plan
- [x] Dry run correctly identifies 18 previously-missed local branches
- [x] Live run successfully deleted all 18
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/248
## Summary
- New `mise run branch-cleanup` task that finds branches merged into main and deletes them locally and on the Forgejo remote
- Configurable `--cutoff` (default 30 days) skips branches with recent HEAD commits
- Supports `--dry-run`, `--local-only`, `--remote-only` flags
- Interactive confirmation before any deletion
## Test plan
- [x] `mise run branch-cleanup -- --dry-run` shows correct table of candidates
- [ ] Run without `--dry-run` to confirm actual deletion works
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/247
## Summary
- Replace bash `indri-runner-logs` with a Python Typer CLI `runner-logs` that supports filtering by runner host (`indri`, `ringtail`, or `all`) with rich table output
- Add missing `#USAGE` declarations to `docs-review`, `docs-review-stale`, and `service-review` so flags work without the `--` separator
- Update docs references in `review-documentation.md` and `review-services.md` to use the new flag syntax
## Test plan
- [x] `mise run runner-logs all` lists runs from both runners
- [x] `mise run runner-logs ringtail` filters to ringtail-only runs
- [x] `mise run docs-review-stale --threshold 90` works without `--`
- [x] `mise run docs-review --limit 5` works without `--`
- [x] `mise run service-review --limit 3` works without `--`
- [x] Pre-commit hooks pass
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/244
## Summary
- Switch from MQTT to webapi polling (v0.5.4 requires only one method)
- Poll every 15s for responsive alerts
- **`notify_once: true`** — one notification per event instead of repeats as object changes zones
- **`nosnap: drop`** — skip events without snapshots (was causing all events to be dropped on v0.3.5)
- **`snap_hires: true`** — use recording stream for higher quality snapshot images
## Deployment and Testing
- [ ] Sync: `argocd app set frigate --revision fix/frigate-notify-config && argocd app sync frigate`
- [ ] Verify pod starts: `kubectl --context=k3s-ringtail -n frigate get pods -l app=frigate-notify`
- [ ] Check logs for successful startup and event processing (no "No snapshot" drops)
- [ ] Wait for a motion event and confirm single ntfy notification with hi-res snapshot
- [ ] After merge: `argocd app set frigate --revision main && argocd app sync frigate`
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/242
Deploy branding.xml with a "Sign in with Authentik" button in the
login disclaimer. Local password login remains available.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Five container manifests were removed when deleting old-style tags
(shared digests). Rebuild on a72a0d8 and update references.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Updates all 15 container image references across 14 ArgoCD manifest files
- Migrates from old internal `vX.Y.Z` tags to new `v<upstream-version>-<sha>` format
- Covers: authentik, cv, devpi, forgejo-runner, homepage, kiwix-serve, kubectl, miniflux, navidrome, ntfy, quartz, teslamate, transmission
## Deployment and Testing
- [ ] Sync all ArgoCD apps on branch revision
- [ ] Verify all services come up healthy
- [ ] Merge and re-sync on main
- [ ] Clean up old-style tags from zot registry
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/238
- harden-zot-registry: fix Authentik hostname, check off all
verified items, add metrics config to "what was done"
- enforce-tag-immutability: fix admins permissions (was missing
update)
- agent-change-process: clarify that requires: is permanent and
status: active is the only completion marker
- zot reference: update modified date
- wire-ci-registry-auth fragment: add metrics fix
- Remove stale harden-zot-mikado-cards.ai.md planning fragment
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add accessControl.metrics.users with empty string to allow
unauthenticated Prometheus/Alloy scraping. Zot represents
anonymous users with an empty username internally.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Enable OIDC + API key authentication on zot registry with three-tier accessControl
- `anonymousPolicy: ["read"]` — anyone can pull
- `artifact-workloads` group: `["read", "create"]` — CI push, no overwrite/delete
- `admins` group: `["read", "create", "update", "delete"]` — break-glass
- Wire both CI push paths (Dagger and Nix/skopeo) with `ZOT_CI_API_KEY` credentials
- Add `artifact-workloads` PolicyBinding in Authentik blueprint for zot app access
- Add `ZOT_CI_API_KEY` to Forgejo Actions secrets via existing ansible role
Completes the `wire-ci-registry-auth` and `harden-zot-registry` Mikado cards.
## Manual Deployment Steps (after merge)
1. Deploy Authentik blueprint: `argocd app sync authentik`
2. In Authentik admin UI: set a password for the `zot-ci` service account
3. Deploy zot config: `mise run provision-indri -- --tags zot`
4. Log in to `https://registry.ops.eblu.me` as `zot-ci` via OIDC → generate API key
5. Store API key in 1Password as `zot-ci-apikey` in blumeops vault
6. Sync Forgejo secrets: `mise run provision-indri -- --tags forgejo_actions_secrets`
7. Trigger a test container build to verify CI push
8. Verify anonymous pull: `curl -sf https://registry.ops.eblu.me/v2/_catalog`
## Uncertainties
- **Zot `accessControl` group matching with OIDC:** Groups from Authentik's `profile` scope claim should map to zot policy groups, but the exact claim-to-group matching needs runtime verification
- **`http.auth.apikey: true`:** This config key is documented but needs verification against the specific zot version built from source on indri
- **API key permissions:** Need to confirm zot API keys inherit the generating user's group for accessControl evaluation
## Test Plan
- [ ] `mise run provision-indri -- --check --diff --tags zot` shows expected config changes
- [ ] Anonymous pull works after deploy
- [ ] Unauthenticated push fails (401)
- [ ] OIDC browser login redirects to Authentik and back
- [ ] API key push works after key generation
- [ ] CI push succeeds with both Dagger and skopeo paths
- [ ] `mise run services-check` passes
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/237
Completed in PR #236. Updated card to reflect what was actually
implemented, including deviations (worker env var wiring, manual
service account setup).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Add Authentik blueprint (`zot.yaml`) with OAuth2 provider, application, `artifact-workloads` group, and `zot-ci` service account
- Wire `zot-client-secret` through ExternalSecret → worker Deployment env var → blueprint `!Env`
- Add Ansible pre_task to fetch OIDC secret from 1Password (item ID `oor7os5kapczgpbwv7obkca4y4`)
- Add `oidc-credentials.json.j2` template and deploy task in zot role (with `when` guard)
## Manual Steps Required Before Deploy
1. Generate client secret: `openssl rand -hex 32`
2. Store in 1Password: add field `zot-client-secret` to "Authentik (blumeops)" item in vault `blumeops`
## What This Does NOT Do
- Does NOT modify `config.json.j2` (that's the root goal `harden-zot-registry`)
- Does NOT wire CI auth (that's `wire-ci-registry-auth`)
- Does NOT set service account password or API keys (manual post-deploy)
## Verification
After ArgoCD sync:
- [ ] Authentik admin UI shows "Zot Registry" application
- [ ] OIDC discovery at `https://authentik.ops.eblu.me/application/o/zot/.well-known/openid-configuration` returns valid JSON
- [ ] Blueprint status is `successful`
- [ ] `artifact-workloads` group exists with `zot-ci` service account
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/236
## Summary
- Removed `status: active` from `enforce-tag-immutability` card — its requirements are folded into the parent `harden-zot-registry` goal's `accessControl` configuration
- Updated `harden-zot-registry` with three-tier access control spec (anonymous read, artifact-workloads read+create, admins full)
- Added `artifact-workloads` group creation step to `register-zot-oidc-client`
- Added service account context to `wire-ci-registry-auth`
## Rationale
Tag immutability requires authentication to be meaningful. Without auth, everyone is anonymous and gets the same policy. Rather than client-side push checks, the registry enforces immutability server-side: CI gets `["read", "create"]` (no update/delete), so pushing an existing tag is rejected by zot itself.
## Test plan
- [ ] `mise run docs-check-links` passes
- [ ] `mise run docs-mikado` shows enforce-tag-immutability as resolved
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/235
Dagger can't run on the bare nix runner (needs container runtime).
Used nix eval directly instead.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Dagger CLI needs a container runtime (Docker/containerd) to start its
engine, which the bare nix runner doesn't have. Use nix eval directly
instead — it's already available and more appropriate for a nix host.
Reverts the dagger flake input since it's not usable here.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Dagger was removed from nixpkgs due to trademark concerns. Use the
official dagger/nix flake as a flake input instead, passing the package
through via specialArgs.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Add `dagger` to `hostPackages` for the ringtail nix-container-builder runner
- Needed for `dagger call nix-version` fallback in the nix build workflow (authentik)
- `hostPackages` is scoped to the runner's systemd unit PATH, not system-wide
- Marks `install-dagger-on-nix-runner` Mikado card complete
## Deployment and Testing
- [ ] Merge, then `mise run provision-ringtail`
- [ ] `mise run container-build-and-release authentik` to verify nix build succeeds
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/234
## Summary
- New Mikado card: the ringtail nix-container-builder runner lacks dagger, which the nix workflow needs for `dagger call nix-version` (authentik version extraction fallback)
- Re-opens `adopt-commit-based-container-tags` with this new prerequisite
- All other containers (11 Dockerfile-only, nettest + ntfy with nix) build fine — only authentik's nix build is blocked
## Deployment and Testing
- Docs only, no deployment needed
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/233
## Summary
- Replace git-tag-triggered container builds with path-based triggers on main and workflow_dispatch
- Image tags now encode upstream app version + commit SHA (`vX.Y.Z-<sha>`) for full traceability
- Replace `container-tag-and-release` task with `container-build-and-release` (dispatches workflows via Forgejo API)
- Update dagger `publish()` to accept `commit_sha` parameter
- Update all docs and references to the new workflow
## Deployment and Testing
- [ ] Merge to main
- [ ] `mise run container-build-and-release <name>` for each container to populate new-format tags
- [ ] Verify tags in registry via `mise run container-list`
- [ ] Existing images untouched — old tags remain available
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/232
## Summary
- Enable OIDC + API key authentication on zot with anonymous pull preserved
- Enforce tag immutability for version tags
- Adopt commit-SHA-based container image tagging
Details in the [[harden-zot-registry]] Mikado chain (`mise run docs-mikado harden-zot-registry`).
## Test plan
- [ ] Anonymous pull still works
- [ ] Unauthenticated push fails (401)
- [ ] CI container builds pass with new auth and tagging
- [ ] `mise run services-check` passes
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/231
## Summary
- New Mikado card: `adopt-commit-based-container-tags` — replaces git-tag-triggered container builds with path-based main-branch triggers and manual workflow dispatch
- Image tags become `vX.Y.Z-<sha>` (with `-main` suffix for main branch builds, `-nix` for Nix builds), tying versions to the actual bundled app version and exact source commit
- `container-tag-and-release` mise task to be renamed to `container-build-and-release`, triggering workflow dispatch with the current HEAD SHA
- Added as soft prereq to `harden-zot-registry` Mikado chain
## Test plan
- [x] Pre-commit hooks pass (docs-check-index, docs-check-links, etc.)
- [ ] Review card content for completeness
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/230
## Summary
- Replace the old pre-Mikado plan doc (`docs/how-to/plans/harden-zot-registry.md`) with a proper C2 Mikado chain in `docs/how-to/zot/`
- Root goal: `harden-zot-registry` — enable OIDC + API key auth on zot with anonymous pull preserved
- Three leaf prereqs: `register-zot-oidc-client`, `wire-ci-registry-auth`, `enforce-tag-immutability`
- Add Zot section to `how-to.md` index, remove plan entry from plans index
- All doc checks pass (`docs-check-links`, `docs-check-index`, `docs-mikado`)
## Changes
- **New:** `docs/how-to/zot/harden-zot-registry.md` — C2 Mikado root goal
- **New:** `docs/how-to/zot/register-zot-oidc-client.md` — Register OIDC client in Authentik
- **New:** `docs/how-to/zot/wire-ci-registry-auth.md` — Wire CI push paths with registry auth
- **New:** `docs/how-to/zot/enforce-tag-immutability.md` — Prevent version tag overwrites
- **Deleted:** `docs/how-to/plans/harden-zot-registry.md` — Old plan doc (content absorbed into Mikado cards)
- **Updated:** `docs/how-to/how-to.md` — Add Zot section, remove plan entry
- **Updated:** `docs/how-to/plans/plans.md` — Remove plan entry
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/229
## Summary
C2 Mikado chain for deploying Authentik as the SSO identity provider, replacing Dex.
This PR will evolve over multiple sessions. Each iteration adds documentation (prerequisite cards) and eventually code as leaf nodes are resolved.
## Current Mikado State
- **Goal:** `deploy-authentik` (active)
- **Leaf prerequisites:**
- `build-authentik-container` — Build Nix container image
- `provision-authentik-database` — Create PostgreSQL database on CNPG cluster
- `create-authentik-secrets` — Create 1Password item with credentials
## Process refinements
- Updated agent-change-process with lessons from first attempt: reset code before committing cards, open PRs early
## Test plan
- [ ] `mise run docs-mikado` shows correct dependency chain
- [ ] Leaf nodes can be worked independently
- [ ] Container builds on ringtail
- [ ] Authentik starts and reaches healthy state
- [ ] Forgejo OAuth2 connector works
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/227
## Summary
- Strip detailed phase instructions from deploy-authentik plan (400→50 lines)
- Retain architecture decisions (ringtail, CNPG on indri, Nix containers, kustomize, Tailscale+Caddy) and open questions
- Add `status: active` frontmatter — now visible as a root goal in `mise run docs-mikado`
- Update plans index to reflect Active (C2) status
This is the first real use of the C2 Mikado chain system from #225. Future sessions will discover prerequisites, create sub-cards with `requires`, and work leaf nodes first.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/226
## Summary
- New plan document at `docs/how-to/plans/deploy-authentik.md` covering full Authentik deployment
- 6 phases: Helm analysis, prerequisites (CNPG/Redis/1Password), Nix containers, kustomize manifests, networking, monitoring
- Authentik replaces Dex as the identity provider for central user management and multi-protocol SSO
- Updated plans index and how-to index
## Deployment and Testing
- [x] Pre-commit hooks pass (docs-check-links, docs-check-index, docs-check-frontmatter)
- [ ] Review plan content for accuracy and completeness
- No deployment needed — documentation only
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/224
## Summary
- Create Dex reference card (`docs/reference/services/dex.md`) with quick reference, architecture, identity source, storage, OIDC clients, secrets, and endpoints
- Write federated login explanation article (`docs/explanation/federated-login.md`) covering the Dex + Forgejo two-layer auth model, login flow, and break-glass access
- Add Dex to `services-check` (HTTP health endpoint + k3s pod check)
- Update Grafana docs with new Authentication section documenting SSO via Dex
- Update Forgejo docs with OAuth2 Provider section documenting its role as upstream identity source
- Add Dex to ringtail workloads table and reference service index
- Move `adopt-oidc-provider` plan to `completed/` with final design reflecting actual implementation
## Test plan
- [ ] `mise run services-check` passes (includes new Dex checks)
- [ ] `docs-check-links` passes (all wiki-links resolve)
- [ ] `docs-check-index` passes (new docs are indexed)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/223
## Summary
- Synced driveway_entrance zone coordinates from live Frigate config (adjusted mask boundaries)
- Added `inertia: 3` and `loitering_time: 0` to driveway_entrance zone
- Expanded review alerts to require either `driveway_entrance` or `driveway` zone (was entrance only)
- Updated frigate-notify config to allow alerts from both `driveway_entrance` and `driveway` zones
## Deployment and Testing
- [ ] Merge and sync frigate ArgoCD app on ringtail
- [ ] Sync frigate-notify (restart pod to pick up ConfigMap change)
- [ ] Verify alerts fire for person/car in driveway zone
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/219
## Summary
- Delete `ansible/roles/frigate_detector/` and remove from indri playbook — the Apple Silicon Detector is retired
- Move Mosquitto (MQTT) ArgoCD app from indri minikube to ringtail k3s
- Move ntfy ArgoCD app from indri minikube to ringtail k3s
- Update Frigate docs to reflect detector removal and planned RTX 4080 migration
- Manifests are reused as-is (same `argocd/manifests/mosquitto/` and `argocd/manifests/ntfy/`), just pointed at ringtail
## Deployment
After merge:
1. Sync indri ArgoCD `apps` app with prune to remove old mosquitto/ntfy apps:
```
argocd app sync apps --prune
```
2. Sync new ringtail apps:
```
argocd app sync mosquitto-ringtail
argocd app sync ntfy-ringtail
```
3. Manually clean up the detector LaunchAgent on indri:
```
ssh indri 'launchctl unload ~/Library/LaunchAgents/mcquack.eblume.frigate-detector.plist'
ssh indri 'rm ~/Library/LaunchAgents/mcquack.eblume.frigate-detector.plist'
```
## Notes
- Frigate on indri will lose MQTT/ntfy connectivity — this is expected (user confirmed no downtime concerns)
- ntfy Tailscale Ingress hostname `ntfy` will transfer from indri ProxyGroup to ringtail ProxyGroup
- Caddy on indri proxies `ntfy.ops.eblu.me` → `ntfy.tail8d86e.ts.net`, so no Caddy changes needed
- Frigate + frigate-notify will be ported to ringtail in a follow-up PR
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/216
## Summary
- Add `containers/nettest/default.nix` using `dockerTools.buildLayeredImage` with curl, jq, dnsutils, cacert, and bash — equivalent to the existing Dockerfile
- Update `container-tag-and-release` to require `--nix` or `--dockerfile` flag when both build types exist for a container
- Update `container-list` to show `[dockerfile+nix]` label when both exist
## Deployment and Testing
- [ ] SSH to ringtail, run `nix build -f containers/nettest/default.nix -o result` to verify the nix expression builds
- [ ] Tag `nettest-nix-v1.0.0`, confirm `build-container-nix` workflow runs on `nix-container-builder` runner and pushes to registry
- [ ] Smoke test on ringtail k3s: `kubectl run nettest --image=registry.ops.eblu.me/blumeops/nettest:v1.0.0 --restart=Never && kubectl logs nettest`
- [ ] Verify `mise run container-list` shows `[dockerfile+nix]` for nettest
- [ ] Verify `mise run container-tag-and-release nettest v1.1.0` prompts for build type
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/214
## Summary
- Replace `changed_when: true` with `register` + output inspection on the two 1Password secret tasks in `ringtail.yml`
- Tasks now correctly report `ok` when the secret content hasn't changed, and `changed` only when `kubectl apply` outputs `configured` or `created`
## Test plan
- [ ] Run `mise run provision-ringtail` twice — second run should show both tasks as `ok` not `changed`
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/213
## Summary
- Adds `inhibit_idle fullscreen` window commands to sway config on ringtail
- Covers both Wayland-native (`app_id`) and XWayland (`class`) windows
- Prevents swayidle from locking the screen during gamepad-only gaming sessions where controller input isn't detected by the Wayland idle tracker
## Notes
This is a blanket fullscreen inhibit. A more targeted approach (daemon monitoring `/dev/input` gamepad events) may be desired later to allow idle lock during long-running fullscreen apps like Factorio.
## Deployment and Testing
- [ ] `mise run provision-ringtail` to deploy
- [ ] Run a fullscreen app and verify swayidle doesn't lock after 15 minutes
- [ ] Verify lock still activates when no fullscreen window is present
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/212
## Summary
- Configure **swayidle** to lock screen (swaylock) after 15 minutes of inactivity
- Turn off display (DPMS) after 60 minutes, auto-restore on activity
- **swaylock** themed with Catppuccin Macchiato to match existing Sway config
- Add `Mod4+l` keybinding for manual screen lock
- Add PAM service for swaylock authentication
- Disable system suspend/hibernate entirely (workstation should never sleep)
## What changes
All changes in `nixos/ringtail/configuration.nix`:
- `security.pam.services.swaylock` — required for swaylock to authenticate on NixOS
- `systemd.sleep.extraConfig` — blocks all sleep/hibernate modes
- `programs.swaylock` (home-manager) — lock screen appearance config
- `services.swayidle` (home-manager) — idle timeout daemon with lock + DPMS events
- New keybinding `Mod4+l` for manual lock
## Deployment and Testing
- [ ] `mise run provision-ringtail`
- [ ] Verify swayidle is running: `systemctl --user status swayidle`
- [ ] Test manual lock with `Super+l`
- [ ] Verify display DPMS off after idle (can lower timeout temporarily to test)
- [ ] Confirm machine does not suspend: `systemctl status sleep.target`
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/211
k3s kubectl on ringtail needs KUBECONFIG set since the eblume
user doesn't have it in their default environment.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
git ls-remote returns multiple lines when a mirror ref exists
(e.g. refs/remotes/remote_mirror_*/main). Take only the first
line to avoid a false mismatch.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
Extends ringtail from a desktop/gaming NixOS box into an infrastructure node with a k3s cluster, secrets management, and a Forgejo Actions
runner for building containers with Nix.
### K3s cluster
- Single-node k3s with Traefik/ServiceLB/metrics-server disabled (minimal footprint)
- TLS SAN set to `ringtail.tail8d86e.ts.net` so ArgoCD on indri can manage it via Tailscale
- Containerd registry mirrors pull through Zot on indri (`k3s-registries.yaml`)
- Tailscale interface added to `trustedInterfaces` for cross-node ArgoCD access
- `kubectl` added to system packages
### 1Password Connect + External Secrets Operator
- Four new ArgoCD apps targeting `k3s-ringtail`: `1password-connect-ringtail`, `external-secrets-crds-ringtail`, `external-secrets-ringtail`,
`external-secrets-config-ringtail`
- Reuses the same Helm charts/values as indri, just pointed at ringtail's k3s API server
- Bootstrap secrets (`op-credentials`, `onepassword-token`) provisioned by Ansible pre_tasks via `op read`, then applied to the `1password`
namespace in post_tasks
### Systemd Forgejo Actions runner
- Native `services.gitea-actions-runner` with `forgejo-runner` package — no DinD, no k8s pod, runs directly on the NixOS host
- Label `nix-container-builder:host` — jobs execute on the host with `nix`, `skopeo`, `nodejs`, etc. in PATH
- Registration token fetched from 1Password (`Forgejo Secrets/runner_reg`) by Ansible and written to `/etc/forgejo-runner/token.env`
- Runner's dynamic user (`gitea-runner`) added to `nix.settings.trusted-users` for nix daemon access
### Nix container build workflow
- New `.forgejo/workflows/build-container-nix.yaml` triggers on `*-nix-v[0-9]*` tags (e.g. `nettest-nix-v1.0.0`)
- Builds with `nix build -f containers/<name>/default.nix`, pushes to Zot via `skopeo copy`
- Existing Dockerfile workflow guarded with `if: !contains(github.ref_name, '-nix-v')` to avoid double-triggering
### Mise task updates
- `container-tag-and-release` auto-detects `default.nix` vs `Dockerfile` and uses the appropriate tag format (`-nix-v` vs `-v`)
- `container-list` shows build type indicator (`[nix]` / `[dockerfile]`)
## Post-merge
1. `mise run provision-ringtail` — deploys k3s token, runner token, NixOS rebuild
2. Register k3s cluster in ArgoCD (first time only):
```fish
ssh ringtail 'sudo cat /etc/rancher/k3s/k3s.yaml' | \
sed 's|127.0.0.1|ringtail.tail8d86e.ts.net|' > /tmp/k3s-ringtail.yaml
set -x KUBECONFIG /tmp/k3s-ringtail.yaml
argocd cluster add default --name k3s-ringtail
3. Sync ArgoCD apps in order: 1password-connect-ringtail -> external-secrets-crds-ringtail -> external-secrets-ringtail ->
external-secrets-config-ringtail
4. Verify runner: ssh ringtail 'systemctl status gitea-runner-nix-container-builder'
5. Check Forgejo admin panel for ringtail-nix-builder runner online
6. Test: create containers/<name>/default.nix, tag with <name>-nix-v0.1.0
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/209
nixos-rebuild can dirty the tree (e.g. flake.lock updates), which
blocks the Ansible git module. Force ensures we always reset to the
upstream state.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
NixOS doesn't include Python by default. Ansible needs it on the
managed host for module execution.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Sway/wlroots refuses to start on proprietary NVIDIA by default.
Add --unsupported-gpu flag and disable hardware cursors.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Bump Frigate image from `0.16.4-standard-arm64` to `0.17.0-rc2-standard-arm64`
- Adapt `record` config to 0.17 schema: `retain.days`/`mode: all` → `continuous.days`
- Update service docs and version tracker
This is the first step toward the Apple Silicon ZMQ detector. The existing ONNX detector is kept so we can validate the upgrade independently.
## What is NOT changing
- Detector config (still `type: onnx` with YOLO-NAS-s)
- go2rtc streams, MQTT, cameras, zones, review rules
- frigate-notify, storage PVs, Grafana dashboard
## Deployment and Testing
- [ ] `argocd app set frigate --revision upgrade-frigate-0.17 && argocd app sync frigate`
- [ ] Pod starts, `/api/version` returns `0.17.0-rc2`
- [ ] No config errors in pod logs
- [ ] Frigate web UI loads at `https://nvr.ops.eblu.me`
- [ ] Live view works, detection running (`/api/stats` shows `detection_fps > 0`)
- [ ] Recordings being created (`/api/recordings/summary`)
- [ ] MQTT events flowing (check frigate-notify logs)
- [ ] After merge: `argocd app set frigate --revision main && argocd app sync frigate`
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/205
## Summary
- Cap detect FPS to 2 to prevent recording segment backlog from ONNX inference bottleneck (~750ms/frame on ARM64 CPU)
- Sync motion masks from live config (added second mask area)
- Update driveway_entrance zone coordinates from live config
- Add explicit alert labels `[person, car]` while keeping `required_zones: [driveway_entrance]`
## Context
The "No frames have been received" error on the gablecam live view was caused by the detect stream falling behind — ONNX YOLO-NAS-s takes ~750ms per inference on ARM64 CPU, but the sub-stream sends 5 FPS. This caused recording segments to pile up and the ffmpeg watchdog to repeatedly kill/restart the process, creating gaps in the live view.
## Test plan
- [ ] Sync ArgoCD `frigate` app to branch and verify pod restarts cleanly
- [ ] Check `/api/stats` — `skipped_fps` should drop significantly, `process_fps` should be close to 2
- [ ] Verify live view at https://nvr.ops.eblu.me/#gablecam no longer shows "No frames" error
- [ ] Verify detections and alerts still work in the driveway_entrance zone
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/204
## Summary
- Bump External Secrets Operator Helm chart from `helm-chart-1.3.1` to `helm-chart-2.0.0` (operator v1.3.2)
- Updates both the operator app and CRDs app `targetRevision`
- No Helm values changes needed — `installCRDs`, `resources`, `webhook`, `certController` keys are unchanged
## Breaking changes in chart 2.0.0
- **Removed providers:** Alibaba and Device42 (unmaintained) — does not affect our 1Password setup
- **Templating engine v1 deprecated** — our ExternalSecrets don't set `engineVersion`, so they use the default (v2)
- **Webhook `failurePolicy`** for SecretStore is now dynamic
## Deployment
1. Sync CRDs first: `argocd app set external-secrets-crds --revision update/external-secrets-helm-2.0.0 && argocd app sync external-secrets-crds`
2. Sync operator: `argocd app set external-secrets --revision update/external-secrets-helm-2.0.0 && argocd app sync external-secrets`
3. Verify: `kubectl --context=minikube-indri -n external-secrets get pods`
4. After merge, set both apps back to `--revision main`
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/203
## Summary
- Add `containers/ntfy/Dockerfile` — three-stage build (Node web UI, Go+CGO server, Alpine runtime) pinned to commit SHA `a03a37fe` (v2.17.0), sourced from forge mirror
- Update ntfy deployment image from `binwiederhier/ntfy:v2.17.0` to `registry.ops.eblu.me/blumeops/ntfy:v1.0.0`
- Note fish shell in CLAUDE.md
## Deployment
After merge, release the container image:
```fish
mise run container-tag-and-release ntfy v1.0.0
```
Then sync:
```fish
argocd app sync ntfy
```
## Test plan
- [x] `docker build` succeeds
- [x] `dagger call build --src=. --container-name=ntfy` succeeds (exit 0, container ID printed)
- [x] `ntfy --help` works in built container
- [ ] Tag and release `ntfy-v1.0.0` after merge
- [ ] Verify ntfy pod starts with new image
- [ ] Verify health endpoint responds at `ntfy.ops.eblu.me/v1/health`
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/202
## Summary
- Upgrade ntfy from v2.11.0 to v2.17.0 (6 minor releases, no breaking changes)
- Add reference doc for ntfy service
- Add reference doc for frigate service (ntfy's sole producer via frigate-notify)
- Update reference index and service-versions.yaml tracking
## Notable upstream changes (v2.12.0–v2.17.0)
- **v2.14.0:** Declarative users/ACL config in files
- **v2.15.0:** `require-login` flag for topic-level auth
- **v2.16.0:** Dead man's switch (heartbeat) notifications, notification update/delete
- **v2.17.0:** Priority templating, crash fixes (nil pointer panics)
## Deployment and Testing
- [ ] ArgoCD sync ntfy after merge
- [ ] Verify ntfy pod healthy with new image
- [ ] Send a test notification via `curl -d "test" https://ntfy.ops.eblu.me/test`
- [ ] Verify frigate-notify still delivers alerts to ntfy
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/201
## Summary
- **Doc review:** Reviewed `gandi-operations.md` — added `last-reviewed` frontmatter, verified all wiki-links, confirmed Pulumi state has no drift
- **Gandi reference fix:** Added missing `cv.eblu.me` CNAME row to `gandi.md` DNS records table (was present in Pulumi but undocumented)
- **Pulumi comment fix:** Updated stale `README.md` reference in `__main__.py` to point to `docs/how-to/gandi-operations.md`
- **How-to reorg:** Moved 14 how-to guides into 3 subdirectories (`deployment/`, `configuration/`, `operations/`), collapsed the Documentation and Database index sections into Configuration and Operations respectively
## Verification
- `docs-check-links` — all 180 wiki-links valid
- `docs-check-filenames` — all 90 filenames unique
- `dns-preview` — 5 resources unchanged, no drift
- All pre-commit hooks pass
## Test plan
- [ ] Verify docs site builds correctly with new paths
- [ ] Spot-check a few wiki-links from other pages to moved how-to guides
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/200
## Summary
- Added a new `build_quartz` Dagger function that builds the Quartz site from a pre-processed source tree (no towncrier)
- Reordered the release workflow so towncrier runs **once** on the runner, then passes the updated working tree to `build-quartz`
- `build_docs` and `build_changelog` are preserved for standalone use — `build_docs` now delegates to `build_quartz` internally
## Motivation
Previously towncrier ran twice per release: once inside a Dagger container (via `build_docs` → `build_changelog`) and once on the runner to capture CHANGELOG.md changes for the git commit. This was wasteful and fragile — if towncrier behavior changed, the two runs could produce different results.
## Test plan
- [ ] Review diff to confirm workflow step ordering is correct
- [ ] Trigger a release and confirm towncrier runs only once
- [ ] Verify the docs tarball contains the updated CHANGELOG.md
- [ ] `dagger call build-quartz --src=. --version=vX.Y.Z` should work standalone
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/199
## Summary
- The `eclipse-mosquitto:2.1.2` tag doesn't exist on Docker Hub — the 2.1.x series only publishes `-alpine` variants
- Corrects the pinned tag to `2.0.22`, the latest non-alpine version (matching what the old floating `:2` tag was resolving to)
- Updates tracking file and changelog fragment accordingly
## Context
The previous PR #197 pinned mosquitto from floating `:2` to `2.1.2`, but the new pod failed with `ErrImagePull` ("manifest unknown"). The old pod is still running on `:2`.
## Test plan
- [ ] Verify `eclipse-mosquitto:2.0.22` pulls successfully
- [ ] Verify mosquitto pod restarts and passes readiness/liveness probes
- [ ] `mise run services-check` passes
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/198
## Summary
- Add `service-versions.yaml` tracking file with 33 services and upstream release URLs
- Add `mise run service-review` task (Python uv script) mirroring the docs-review UX
- Add `review-services` how-to article covering the review process by service type
- Add `[[review-services]]` link to the how-to index Knowledge Base table
## Deployment and Testing
- [x] `mise run service-review` displays 33 services, all "never reviewed"
- [x] `mise run service-review -- --type ansible` filters to 7 Ansible services
- [x] `mise run service-review -- --limit 5` shows 5 rows
- [x] `mise run docs-check-links` — no broken wiki-links
- [x] `mise run docs-check-frontmatter` — new doc passes validation
- [x] All pre-commit hooks pass
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/196
## Summary
- Switch navidrome deployment from upstream `deluan/navidrome:0.60.3` back to custom image `registry.ops.eblu.me/blumeops/navidrome:v1.0.2`
- The v1.0.1 image was tagged before the `USER 65534` removal commit, so it still ran as a non-root user that couldn't write to the SQLite data directory
- v1.0.2 is built from current main which includes both the `zlib-dev` build fix and the non-root user removal
## Deployment and Testing
- [ ] Wait for CI to build `navidrome:v1.0.2` image
- [ ] Sync via ArgoCD and verify pod starts without CrashLoopBackOff
- [ ] Verify navidrome UI accessible at https://navidrome.ops.eblu.me
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/194
## Summary
- Remove car-specific `max_frames: 150` which was causing a forget-and-re-detect loop on parked cars (every ~30 seconds at 5fps)
- Set `stationary.interval: 0` so Frigate never re-runs detection on stationary objects
- Replace read-only configmap subPath mount with initContainer + emptyDir, so Frigate UI changes (zones, masks) persist at runtime
## Context
Frigate was spamming notifications because `max_frames` for cars caused it to "forget" a parked car after 150 frames, then immediately re-detect it as a brand new object. The fix follows [Frigate's official parked cars guide](https://docs.frigate.video/guides/parked_cars/).
The writable config change also unblocks using `required_zones` for car alerts — zones can now be drawn in the Frigate UI and will survive until pod reschedule (at which point they should be baked into the configmap via IaC).
## Test plan
- [ ] Sync frigate app via ArgoCD and verify pod starts with initContainer
- [ ] Confirm parked cars no longer trigger repeated alerts
- [ ] Draw a zone/mask in Frigate UI, save, verify it persists after Frigate restart
- [ ] Set up `driveway_entrance` required zone for car alerts
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/193
## Summary
- Add how-to guide (`docs/how-to/build-container-image.md`) covering the full container build workflow: directory layout, Dagger local builds, mise release task, and common patterns with links to existing containers
- Port navidrome from upstream `deluan/navidrome:0.60.3` to a custom three-stage build (`containers/navidrome/Dockerfile`) using Node + Go + Alpine
- Update navidrome deployment to use `registry.ops.eblu.me/blumeops/navidrome:v1.0.0`
## Deployment and Testing
- [x] `dagger call build --src=. --container-name=navidrome` builds successfully
- [ ] After merge: `mise run container-tag-and-release navidrome v1.0.0`
- [ ] After image published: `argocd app sync navidrome` and verify pod starts
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/192
## Summary
Review session covering 3 docs, plus a codebase-wide cleanup:
### Docs reviewed
- **connect-to-postgres** — verified end-to-end (psql connection tested), stamped
- **create-release-artifact-workflow** — clarified that `build-blumeops.yaml` is only a version bump example (not a packages API example)
- **deploy-k8s-service** — fixed stale repoURL (`indri:2200` → `forge.ops.eblu.me:2222`), wrong Caddy config keys (`upstream` → `backend`, added missing `host`), updated Homepage group to "Services", added Tailscale tag documentation
### Codebase cleanup
- Migrated all remaining `op item get --fields` calls to `op read` URI syntax across 7 files (docs, READMEs, YAML comments)
- Simplified the `op read` vs `op item get` guidance in CLAUDE.md
## Side findings (not addressed)
- New `immich-pg` CNPG cluster not yet documented in the postgresql reference card
## Test plan
- [x] `psql` connection to `pg.ops.eblu.me` verified
- [x] All pre-commit hooks pass
- [x] `docs-check-links`, `docs-check-index`, `docs-check-frontmatter` pass
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/191
Adds docs/index.md, explanation/explanation.md, how-to/plans/plans.md,
and how-to/plans/completed/completed.md so AI sessions get full
awareness of all doc sections and in-flight plans.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Reflect actual UX7 zone-based firewall UI, correct streaming port
(8096 not 443), note indri DHCP reservation, mark plan as completed.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Abandon the UniFi Pulumi IaC approach after provider bugs caused a network outage (no-op update reset undeclared properties on the default LAN network)
- Remove untracked IaC artifacts (`pulumi/unifi/`, `mise-tasks/unifi-preview`, `mise-tasks/unifi-up`) locally
- Mark `add-unifi-pulumi-stack` plan as Abandoned with explanation
- Create new `segment-home-network` plan for manual three-network segmentation (Main/IoT/Guest) via UX7 web UI
- Rewrite UniFi reference card to remove all Pulumi/IaC references
- Update plan and how-to indexes
## Test plan
- [x] `docs-check-links` passes
- [x] `docs-check-index` passes
- [x] Pre-commit hooks pass
- [ ] Review segmentation plan for completeness before executing manually
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/189
## Summary
- Add new how-to guide (`connect-to-postgres.md`) with the `psql` command using `op read` for 1Password credentials
- Add "Database" section to the how-to index linking to the new guide
- Link the new guide from the PostgreSQL reference card's Related section
## Test plan
- [x] Verified `psql` connection works from gilbert using the documented command
- [ ] Review doc formatting and content
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/188
Chart 2.3.0 mounts credentials as a file with standard k8s base64
encoding. The old double-encoding workaround (credentials-base64 in
stringData) now produces invalid JSON. Use raw JSON (credentials-file)
instead.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Replace `op item get --fields` with `op read` for secrets (matches playbook and CLAUDE.md guidance)
- Change `tags: [<role>]` to `tags: <role>` to match actual playbook style
- Remove redundant `listen:` from handler example, add `changed_when: true`
- Name handler after specific service (e.g. `Restart <service>`) to match real roles
- Add `last-reviewed: 2026-02-13` frontmatter
## Also noted (not fixed here)
Two other docs still use the old `op item get` pattern:
- `docs/how-to/troubleshooting.md:72` (ArgoCD login command)
- `docs/how-to/gandi-operations.md:35` (Gandi token export)
These can be fixed in their own review cycles.
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/185
## Summary
- Add `daemon.json` with `registry-mirrors` to the forgejo-runner ConfigMap, pointing DinD at `http://host.minikube.internal:5050`
- Mount `daemon.json` into the DinD sidecar at `/etc/docker/daemon.json` via `subPath`
- Docker Hub pulls during Dagger CI builds will now route through Zot's pull-through cache, reducing bandwidth and avoiding rate limits
## Deployment and Testing
- [ ] `argocd app sync forgejo-runner`
- [ ] Exec into DinD container: `docker info` should show the registry mirror
- [ ] Trigger a workflow build and check Zot logs for cache hits
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/183
v3.2.0 build failed (GitHub download timeout), rolling back to
working image while it rebuilds.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Move FORGEJO_URL, RUNNER_NAME, and RUNNER_LABELS from ExternalSecret template to deployment env vars
- ExternalSecret now only contains the actual secret (RUNNER_TOKEN)
- Image version changes in RUNNER_LABELS now trigger automatic pod rollouts
## Deployment
1. Merge this PR
2. `argocd app sync forgejo-runner` — the deployment spec change will auto-roll the pod
No manual restart needed — that's the whole point :)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/181
## Summary
- Install yq in the forgejo-runner container image for structured YAML editing
- Replace fragile `sed` regex patterns with `yq` in `build-blumeops.yaml` and `cv-deploy.yaml` workflows
## Deployment
1. Merge this PR
2. Tag and release forgejo-runner v3.1.0: `mise run container-tag-and-release forgejo-runner v3.1.0`
3. Update runner label in `argocd/manifests/forgejo-runner/external-secret.yaml` from `v3.0.2` to `v3.1.0`
4. Sync the forgejo-runner app: `argocd app sync forgejo-runner`
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/180
Replace old Apps/Observability/Infrastructure layout entries with
Content and Misc to match the recategorized ingress annotations.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
ArgoCD's tailscale ingress was missed in the recategorization (filed as
service-tailscale.yaml instead of ingress-tailscale.yaml). Fix the group
annotation and rename the file to match the convention used by all other
services.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Replace the three homepage groups (Apps, Observability, Infrastructure) with two cleaner groups
- **Content**: Immich, Kiwix, Miniflux, DJ, Grafana
- **Misc**: CV, TeslaMate, Transmission, Docs, Prometheus, PyPI
## Deployment and Testing
- [ ] Sync affected ingresses via ArgoCD (all 11 services)
- [ ] Verify homepage shows the two new groups correctly
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/179
## Summary
- Remove `match_all = true` from `flyio_nginx_cache_requests_total` in Alloy so the metric only counts requests that go through the proxy cache (excludes health checks with empty `cache_status`)
- Change dashboard queries from `rate(...[5m])` to `increase(...[$__range])` — aggregates over the full dashboard time window instead of a 5-minute sliding window, giving meaningful ratios for low-traffic static sites
- Add null/NaN value mapping to show "No traffic" in neutral color instead of blank/red
## Root cause
Health check requests from Fly.io hit the default nginx server block (no `proxy_cache`), producing entries with empty `upstream_cache_status`. With `match_all = true`, these were counted in the cache metric, diluting the Fly.io dashboard ratio. For APM dashboards, `rate()[5m]` on low-traffic sites with 24h cache validity almost always returns either all-HITs (100%) or no data (blank → red background).
## Deployment
- Fly.io proxy redeploy needed for Alloy config change
- ArgoCD sync for dashboard ConfigMap changes
## Test plan
- [ ] Redeploy Fly.io proxy
- [ ] Sync grafana-config in ArgoCD
- [ ] Verify CV APM cache hit ratio shows a real percentage (not 100%)
- [ ] Verify Docs APM shows "No traffic" in neutral color when idle, real ratio when visited
- [ ] Verify Fly.io proxy dashboard cache ratio excludes health checks
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/177
## Summary
- How-to guide for creating release artifact workflows with Forgejo packages
- Changelog fragment for the multi-repo forgejo_actions_secrets Ansible role change
- Changelog fragment for the new docs
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/170
Uses subelements loop to sync secrets across repos. Adds FORGE_TOKEN
to the cv repo for package uploads.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- nginx container (`containers/cv/`) downloads and serves a content tarball at startup (same pattern as quartz)
- ArgoCD app + k8s manifests (deployment, service, Tailscale ingress)
- Caddy route for `cv.ops.eblu.me`
- Deploy workflow: resolves "latest" or specific version from Forgejo packages, updates deployment, syncs ArgoCD
- Content is built and released from the separate [cv repo](https://forge.ops.eblu.me/eblume/cv)
## Deployment steps (after merge)
1. `mise run container-tag-and-release cv v1.0.0`
2. Run "Release CV" workflow in cv repo (SPECIFIC_VERSION `v0.1.0`)
3. Run "Deploy CV" workflow in blumeops (default: latest)
4. `mise run provision-indri -- --tags caddy`
5. Verify at `https://cv.ops.eblu.me/`
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/169
## Summary
- The `build_changelog` Dagger container (`python:3.12-slim`) defaults to UTC, causing towncrier to stamp tomorrow's date when releases are cut in the evening PST.
- This is the root cause of the docs website (built via Dagger) showing Feb 12 while the repo CHANGELOG (built directly on the runner) correctly showed Feb 11.
- Fix: set `TZ=America/Los_Angeles` in the Dagger container before running towncrier.
## Verified
- `docker run --rm python:3.12-slim` → `towncrier _get_date()` returns `2026-02-12` (wrong)
- `docker run --rm -e TZ=America/Los_Angeles python:3.12-slim` → returns `2026-02-11` (correct)
## Test plan
- [ ] Merge, then trigger a build-blumeops release
- [ ] Verify the CHANGELOG date on https://docs.eblu.me/CHANGELOG matches the repo
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/167
## Summary
- Updated the "Configure Ingress" section to use the current ProxyGroup pattern (`proxy-group: "ingress"`, `defaultBackend`, `tls.hosts`)
- Replaced the old per-ingress proxy example that used `rules:` with `host:` (which breaks ProxyGroup routing)
- Added key points explaining why `defaultBackend` is required and what each annotation does
- Updated checklist to mention ProxyGroup
## Test plan
- [ ] Review rendered doc for accuracy against existing ingress manifests
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/166
## Summary
- Move `adopt-dagger-ci.md` to new `docs/how-to/plans/completed/` archive
- Update plan status to "Phases 1–3 complete" with all verification checklists checked
- Update runner description to reflect what actually stayed (Docker CLI, Node.js needed)
- Document known issue: changelog dates still show UTC
- Create completed plans index, update plans and how-to indexes
- Changelog fragment
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/165
## Summary
- Dagger shells out to the `docker` binary to provision its BuildKit engine container
- Phase 3 removed `docker-ce-cli`, breaking all `dagger call` invocations in CI
- This restores `docker-ce-cli` (without buildx/skopeo — those aren't needed)
## Test plan
- [ ] Build locally, release as v3.0.2, update manifest, sync
- [ ] Trigger docs build workflow and verify Dagger engine starts
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/164
## Summary
- Restores Node.js 20 LTS to the Forgejo runner job image
- `actions/checkout@v4` and other JavaScript Actions require `node` in the job container
- The Phase 3 simplification (PR #162) accidentally removed it, breaking all CI runs
## Changes
- `containers/forgejo-runner/Dockerfile`: Add `gnupg` (for nodesource GPG key) and Node.js 20 via nodesource
- Changelog fragment
## Test plan
- [ ] Merge, release as `forgejo-runner-v3.0.1`
- [ ] Update runner manifest to v3.0.1, sync, restart pod
- [ ] Trigger a workflow_dispatch and verify `actions/checkout` succeeds
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/163
## Summary
With Phases 1 and 2 complete, the runner image no longer needs most of its bundled tools. This PR strips it down and adds what was missing.
**Removed** (now inside Dagger containers):
- Node.js 24.x
- Docker CLI + buildx plugin
- skopeo
- gnupg, lsb-release, xz-utils
**Added:**
- `tzdata` — fixes the TZ env var (#159, #160, #161) so `TZ=America/Los_Angeles` actually works
- `flyctl` — was being installed from scratch every release
**Workflow changes:**
- Remove "Ensure Dagger CLI" bootstrap steps from both workflows (Dagger is in the image)
- Remove "Install flyctl" step from build-blumeops (flyctl is in the image)
- Remove job-level `TZ` from build-blumeops (moved to runner configmap `runner.envs`)
- Set `TZ: America/Los_Angeles` in runner configmap so all job containers inherit it
## Deployment
After merge:
1. Build and release the new runner image: `mise run container-release forgejo-runner v2.0.0`
2. Sync the runner: `argocd app sync forgejo-runner`
3. Verify: `kubectl -n forgejo-runner exec deploy/forgejo-runner -c runner -- date` (but the real test is running a docs release and checking the changelog date)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/162
## Summary
The runner pod's `TZ` env var (#159, #160) doesn't propagate to workflow job containers — jobs run inside Docker containers spawned by the DinD sidecar, not in the runner process itself. Set `TZ: America/Los_Angeles` at the job level so `uvx towncrier build` uses the correct timezone.
This is the actual fix for the Feb 12 changelog dates. The runner pod TZ is still useful for runner daemon logs but doesn't affect job execution.
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/161
## Summary
The `TZ=America/Los_Angeles` env var from #159 has no effect because the `forgejo/runner` image doesn't ship tzdata. Mount the node's `/usr/share/zoneinfo` into the container so the timezone database is available.
## Deployment
After merge, sync forgejo-runner and verify:
```
argocd app sync forgejo-runner
kubectl -n forgejo-runner exec deploy/forgejo-runner -c runner -- date
# Should show PST/PDT, not UTC
```
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/160
## Summary
- Set `TZ=America/Los_Angeles` on the Forgejo runner container
The runner pod defaults to UTC. When releases are cut in the evening PST, towncrier stamps changelog entries with tomorrow's date (e.g., v1.6.2 shows 2026-02-12 despite being released on the evening of Feb 11 PST).
## Deployment
After merge, sync the forgejo-runner ArgoCD app:
```
argocd app sync forgejo-runner
```
The runner pod will restart with the new timezone. Note: the v1.6.2 changelog entry will remain dated 2026-02-12; future entries will use PST dates, so dates may appear non-sequential once.
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/159
## Summary
- Rename `date-modified` -> `modified` in all 80 docs and the `docs-check-frontmatter` task
Quartz's `CreatedModifiedDate` plugin recognizes `modified`, `lastmod`, `updated`, and `last-modified` — but not `date-modified`. The wrong field name caused Quartz to ignore frontmatter dates entirely and fall through to filesystem timestamps (UTC inside Dagger), showing Feb 12 on pages built late on Feb 11 PST.
## Test plan
- [x] `mise run docs-check-frontmatter` passes
- [ ] Kick off docs release after merge — verify rendered dates match frontmatter values
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/158
## Summary
Migrates the docs build pipeline to Dagger (Phase 2 of the Dagger CI adoption plan).
- **Backfill `date-modified` frontmatter** on all 80 docs — Dagger's `--src=.` excludes `.git`, so Quartz can't use git history for page dates. Frontmatter dates work with or without git.
- **New `docs-check-frontmatter` mise task + pre-commit hook** — validates all docs have `title`, `tags`, and `date-modified`
- **New Dagger functions** — `build_changelog` (towncrier in Python container) and `build_docs` (chains changelog → Quartz build in Node container, returns tarball)
- **Simplified CI workflow** — the ~44-line inline Quartz build (clone, npm ci, build, tar, cleanup) is replaced by `dagger call build-docs`. Changelog step remains local on the runner since towncrier needs to modify the host working tree for the git commit.
### Design decisions
- **Towncrier runs twice in CI**: once inside Dagger (for the docs tarball) and once on the runner (for the git commit). This is intentional — Dagger's directory export is additive and can't delete the consumed changelog fragments from the host.
- **Artifact hosting stays on Forgejo Releases** (not migrated to Forgejo Packages as the plan doc originally suggested). That migration can happen independently.
- **`date-modified` frontmatter** preserved even though `build_changelog` installs git — the git there is only for towncrier's `git add` call, not for history. The local iteration story (`dagger call build-docs --src=. --version=dev` with uncommitted changes) depends on frontmatter dates.
### Local iteration
```bash
dagger call build-docs --src=. --version=dev export --path=./docs-dev.tar.gz
tar tf docs-dev.tar.gz | head -20
```
## Deployment and Testing
- [x] `dagger call build-docs --src=. --version=dev` produces valid 1.1MB tarball (149 HTML pages)
- [x] Pre-commit hooks pass (including new `docs-check-frontmatter`)
- [ ] Full `workflow_dispatch` run after merge
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/157
## Summary
- Migrate from deprecated Todoist REST API v2 (`410 Gone`) to new unified API v1
- Add cursor-based pagination for project and task listing endpoints
- Switch 1Password credential retrieval from `op item get --fields` to `op read`
## Testing
- [x] `mise run blumeops-tasks` returns all 9 tasks successfully
- [x] Pre-commit hooks pass
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/155
fly ssh console -C doesn't run through a shell, so && was passed as
literal arguments to rm. Wrap in sh -c to get proper shell parsing.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- The Fly.io nginx proxy caches docs responses for 24h (`proxy_cache_valid 200 1d`)
- After a release, docs.eblu.me kept serving stale content until the cache expired
- This caused v1.5.4 to show v1.5.3 on the CHANGELOG page
- Adds `flyctl` install and `fly ssh console` cache purge steps to the build workflow, running after the ArgoCD deploy completes
## Test plan
- [ ] Next release should show the correct version on docs.eblu.me/CHANGELOG immediately
- [ ] Verify the `fly ssh console` command succeeds in the workflow logs
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/154
## Summary
- Bump Fly.io proxy VM memory from 256MB to 512MB — Alloy was OOM-killed, causing the Grafana Fly.io dashboard to lose metrics
- Fix TruffleHog pre-commit hook to scan only staged changes (`--since-commit HEAD`) instead of full repo history
- Sanitize example credential URL in Reolink camera plan doc
## Deployment and Testing
- [ ] Fly.io deploy triggers automatically on merge (workflow watches `fly/**`)
- [ ] After deploy, verify Alloy is running: `fly ssh console -a blumeops-proxy -C "ps aux"` should show alloy process
- [ ] Grafana Fly.io dashboard should start populating within ~1 minute
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/152
## Summary
Two new plan documents in `docs/how-to/plans/`:
- **adopt-dagger-ci** — Migrate CI/CD build logic from Forgejo Actions YAML to Dagger (Python SDK). Forgejo Actions stays as a thin trigger layer. Covers:
- Container builds with local iteration (`dagger call build ... terminal`)
- Docs builds with Forgejo packages migration (replacing Forgejo releases)
- Runner simplification (only Docker + dagger CLI needed)
- Secrets handling via Dagger's `Secret` type
- Future: forked project builds, Python packages, pre-merge validation
- **upstream-fork-strategy** — Stacked-branch pattern for maintaining forks of upstream projects. Covers:
- Daily automated rebase with conflict detection and issue creation
- Branch model: `upstream/main` → `blumeops` → `feature/*`
- Quartz fork as first instance, enabling `last-reviewed` frontmatter rendering in docs
- Upstream PR path for contributing changes back
## Context
These plans emerged from evaluating alternatives to the GHA ecosystem (BuildKite, Concourse, Earthly) for CI/CD. Dagger was chosen for its local iteration story, Python-native pipelines, and zero-infrastructure requirements. The fork strategy is a prerequisite for customizing Quartz and other upstream tools.
Neither plan is ready for execution yet — they are design documents for future work.
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/150
## Summary
- Switch build action back to `docker buildx build` now that runner v2.5.1 (with `docker-buildx-plugin`) is deployed
## Test plan
- [ ] Merge and tag `nettest-v0.12.0` to verify buildx works end-to-end
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/149
## Summary
- Temporarily revert composite action to `docker build` so we can build the runner image (chicken-and-egg: current runner v2.5.0 doesn't have buildx)
- Bump runner label to `v2.5.1` so after sync the new runner image (with buildx) gets used
## Deployment plan
1. Merge this PR
2. Tag `forgejo-runner-v2.5.1` — builds with legacy `docker build` (one last time)
3. Sync forgejo-runner in ArgoCD to pick up the v2.5.1 label
4. Follow-up PR: switch action back to `docker buildx build`
5. Tag `nettest-v0.12.0` to verify buildx works end-to-end
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/148
## Summary
- Install `docker-buildx-plugin` alongside `docker-ce-cli` in the forgejo-runner image
- Fixes `docker buildx build` failing with "unknown flag: --tag" from #146
## Test plan
- [ ] Merge and release `forgejo-runner-v2.5.1`
- [ ] Update runner configmap/labels if needed to use new image
- [ ] Re-tag `nettest-v0.11.1` (or `v0.12.0`) to verify build-container workflow succeeds
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/147
## Summary
- Replace deprecated `docker build` with `docker buildx build` in the build-push-image composite action
- Remove redundant build/run comments from nettest Dockerfile
## Test plan
- [ ] Merge and tag `nettest-v1.1.0` (or similar) to trigger the build-container workflow
- [ ] Verify the build succeeds without the deprecation warning
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/146
## Summary
- Adds guidance to CLAUDE.md: use `op read` for secret values, `op item get` only for metadata
- Fixes the argocd login example which used `op item get --fields`
- `op item get --fields` wraps multi-line values in quotes, which corrupts keys and other secrets
Discovered while verifying the sifaka borg repokey in 1Password — hashes didn't match until we switched to `op read`.
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/143
## Summary
- Adds BorgBase as a second borgmatic repository for offsite backups (US region, append-only)
- SSH key managed via 1Password, deployed to indri by Ansible
- Borgmatic `ssh_command` configured to use the dedicated BorgBase key
- BorgBase host key pinned in known_hosts via Ansible
## Post-merge deployment steps
1. Provision borgmatic: `mise run provision-indri -- --tags borgmatic`
2. Initialize the BorgBase repo: `ssh indri 'mise x -- borgmatic init --encryption repokey --repository borgbase-offsite'`
3. Export and store the borg repokey: `ssh indri 'borg key export ssh://k04ljcd7@k04ljcd7.repo.borgbase.com/./repo'` → save to 1Password
4. Verify first backup: `ssh indri 'mise x -- borgmatic create --repository borgbase-offsite --verbosity 1'`
## BorgBase setup (already done)
- Account created, API token in 1Password (`borgbase` item in blumeops vault)
- SSH keypair generated, stored in 1Password, public key uploaded to BorgBase (ID: 200815)
- Repository `indri-borgmatic` created (ID: k04ljcd7, US region, append-only, 2-day alert)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/142
## Summary
- New how-to guide at `docs/how-to/restore-1password-backup.md` with step-by-step procedure for extracting and decrypting a 1Password `.1pux` export from borgmatic backup
- **End-to-end verified**: extracted from today's borg archive, decrypted age key with openssl, decrypted .1pux with age → valid 31MB zip with vault data
- Cross-links added from: disaster-recovery, 1password, borgmatic, backups policy, and how-to index
- Updated disaster-recovery.md from TBD stub to include a procedures table
## Deployment and Testing
- [x] Verified full extraction + decryption flow against live borgmatic archive
- [x] `docs-check-links` passes — all wiki-links valid
- [ ] Review guide for clarity and completeness
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/141
## Summary
- Add `docs/how-to/plans/migrate-forgejo-from-brew.md` — full Diataxis-style plan covering background, one-time migration steps, Ansible role changes (with exact code), verification checklist, and future considerations
- Add `docs/how-to/plans/plans.md` — new plans subdirectory index for upcoming migration/transition plans
- Update `docs/how-to/how-to.md` with a Plans section
- Update `docs/tutorials/exploring-the-docs.md` to mention plans in the doc structure table and quick-path sections for Owner and AI audiences
## Test plan
- [x] `docs-check-links` passes
- [x] `docs-check-index` passes
- [x] All pre-commit hooks pass
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/140
## Summary
- Documents a troubleshooting procedure for broken pod networking after unclean shutdown
- During minikube recovery, a stale `1-k8s.conflist` CNI config can override kindnet's `10-kindnet.conflist`, causing new pods to use bridge+firewall networking instead of kindnet's ptp — breaking pod-to-pod communication
- Covers symptoms (DNS failures, liveness probe timeouts), diagnosis steps, and the fix
## Context
Encountered this during the 2026-02-10 power outage. Immich, kiwix, and transmission were all crash-looping for ~8 hours due to the CNI conflict. The minikube ansible role's clean boot detection has been improved (#137) so this may not recur, but the troubleshooting guide is valuable if it does.
## Test plan
- [x] Documentation only — no code changes
- [x] Pre-commit hooks pass
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/139
## Summary
- New `power.md` reference card documenting the full power chain: AC grid → Anker SOLIX F2000 battery → CyberPower CP1000PFCLCD UPS → homelab
- Lists all devices on the UPS (indri, sifaka, UniFi Express 7, Starlink)
- Replaced inline UPS entry on indri card with link to the new power card
- Added power card to reference index
Context: power chain was upgraded — the Anker battery station now sits between grid power and the UPS, providing extended runtime for grid outages.
## Test plan
- [x] docs-check-links, docs-check-index, docs-check-filenames all pass
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/138
## Summary
- After a power loss, minikube's Docker container (host) restarts but kubelet/apiserver remain stopped
- The ansible role's status check used `--format='{{.Host}}'` which only examined the host VM state
- When host=Running but kubelet/apiserver=Stopped, the role skipped `minikube start`
- Fixed to use full `minikube status` exit code (returns non-zero when any component is unhealthy)
- Simplified all downstream conditions to use exit code instead of string matching
## Test plan
- [x] Verified the fix correctly skips `minikube start` when cluster is already fully running
- [x] Pre-commit hooks pass (ansible-lint, yamllint, etc.)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/137
## Summary
- Adds `mise run op-backup` task that encrypts a 1Password .1pux export with `age` using the master password + secret key as passphrase, SCPs to indri for borgmatic pickup, then deletes the plaintext
- Adds `age` to the Brewfile
- Borgmatic already backs up `/Users/erichblume/Documents` on indri, which covers the `1password-backup/` subdirectory — no config change needed
## Disaster recovery
1. Restore borgmatic archive to retrieve the `.age` file
2. Open Emergency Kit from safety deposit box
3. `age --decrypt <file>.age > export.1pux` (passphrase: `{master_password}:{secret_key}`)
4. Open `.1pux` with 1Password or unzip to inspect
## Usage
```
# Export all vaults from 1Password desktop app as .1pux, then:
mise run op-backup ~/Documents/1Password-export.1pux
# Or run without args for interactive prompt:
mise run op-backup
```
## Test plan
- [ ] `brew install age`
- [ ] Export a test vault from 1Password as .1pux
- [ ] Run `mise run op-backup` with the export path
- [ ] Verify encrypted file appears on indri at `~/Documents/1password-backup/`
- [ ] Verify plaintext .1pux is deleted from gilbert
- [ ] Test decryption: `age --decrypt <file>.age > test.1pux` with password:secret_key
- [ ] Verify decrypted .1pux can be opened/unzipped
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/136
## Summary
- Add `smartctl_exporter` Docker container to sifaka for SMART disk health monitoring
- Formalize existing `node_exporter` container under Ansible management
- Route both exporters through Caddy L4 TCP proxy (`nas.ops.eblu.me:9100`, `nas.ops.eblu.me:9633`), replacing the hardcoded LAN IP in Prometheus
- Create "Sifaka Disk Health" Grafana dashboard (health status, temperature, wear indicators, lifetime)
- Introduce `ansible/playbooks/sifaka.yml` and `mise run provision-sifaka` — first Ansible playbook for the NAS
- Shared exporter port variables in `group_vars/all.yml` to avoid duplication between Caddy and sifaka roles
## Prerequisites before deploy
- [ ] Enable SSH on sifaka (DSM Control Panel > Terminal & SNMP)
- [ ] Verify `ssh eblume@sifaka 'docker ps'` works
- [ ] Run `mise run provision-sifaka` to deploy containers
- [ ] Run `mise run provision-indri -- --tags caddy` to add L4 routes
- [ ] `argocd app sync prometheus` + `argocd app sync grafana-config`
## Test plan
- [ ] Verify smartctl_exporter metrics: `curl http://nas.ops.eblu.me:9633/metrics`
- [ ] Verify Prometheus targets page shows both sifaka jobs as UP
- [ ] Verify Grafana "Sifaka Disk Health" dashboard loads with data
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/135
## Summary
- Adds a branded 503 error page served when upstreams are unreachable (indri offline, Tailscale tunnel down, emergency shutoff, etc.)
- Stale cache is still served first when available (`proxy_cache_use_stale` takes priority)
- Test endpoint at `docs.eblu.me/_error` to preview the page without killing upstreams
- `proxy_intercept_errors on` also catches error responses returned by the upstream itself
## Files Changed
- `fly/error.html` — Self-contained error page (dark theme, links to BlumeOps repo)
- `fly/nginx.conf` — `error_page`, `internal` location, `/_error` test location, `proxy_intercept_errors`
- `fly/Dockerfile` — COPY error.html into image
## Test Plan
- [ ] Deploy to Fly.io
- [ ] Visit `docs.eblu.me/_error` to verify the page renders
- [ ] Optionally stop indri/Tailscale to confirm the page shows on real 502/503/504
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/133
## Summary
- Start nginx after Tailscale connects (community best practice for Tailscale sidecars)
- Switch to `bluegreen` deploy strategy — old machine serves until new one is healthy
- Replace top-level `[checks]` with `[[http_service.checks]]` — only service-level checks gate traffic routing ([confirmed by Fly.io staff](https://community.fly.io/t/clarifying-the-types-of-health-checks/20379))
- Remove sentinel file and nginx if-check (no longer needed)
Supersedes the approach in #131 — that helped (502 window dropped from ~30s to ~3s) but couldn't fully eliminate it because top-level checks don't gate routing and Fly.io's proxy sends traffic as soon as the port is reachable.
## Deployment and Testing
- [ ] Merge and `fly deploy` from `fly/` directory
- [ ] Verify deploy completes with zero 502s (watch `fly logs` and Grafana docs-apm)
- [ ] Confirm `fly checks list` shows the new service-level check passing
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/132
## Summary
- Health check (`/healthz`) now returns 503 until Tailscale is connected
- `start.sh` creates `/tmp/tailscale-ready` sentinel after `tailscale up` succeeds
- Fly.io keeps the old machine serving traffic during the ~7s startup window
Previously, nginx passed the health check immediately, Fly.io routed traffic to the new machine, but MagicDNS wasn't available yet — causing upstream DNS timeouts and 502s on every request until Tailscale connected.
## Deployment and Testing
- [ ] Merge and `fly deploy` from `fly/` directory
- [ ] Verify deploy completes with zero 502s (check Grafana docs-apm dashboard)
- [ ] Confirm health check transitions from 503 → 200 in `fly logs`
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/131
## Summary
- Add `client_ip` field to the Fly.io nginx JSON log format, sourced from `Fly-Client-IP` header
- Extract `client_ip` in the Alloy pipeline so it's available as a parsed field in Loki
- Keeps `remote_addr` (the internal proxy IP) for debugging
Fixes: Grafana access logs for docs.eblu.me showing 172.16.11.178 for every request instead of real visitor IPs.
## Deployment and Testing
- [ ] Deploy updated fly.io proxy: `fly deploy` from `fly/` directory
- [ ] Verify in Grafana that new log lines include `client_ip` with real IPs
- [ ] Confirm `remote_addr` still shows the proxy IP (preserved for debugging)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/130
## Summary
- New `docs-review` mise task replaces `docs-review-random` — sorts docs by `last-reviewed` frontmatter field (never-reviewed first, then oldest)
- Updated review-documentation how-to to explain the new workflow and how to mark cards as reviewed
- Updated ai-assistance-guide task table to reference `docs-review`
## Test plan
- [x] `mise run docs-review` runs and shows staleness table + most stale doc
- [x] `mise run docs-review -- --limit 5` respects the limit flag
- [x] All pre-commit checks pass (links, index, filenames)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/129
## Summary
- Start nginx before Tailscale in `start.sh` so port 8080 is bound immediately, eliminating the "app is not listening on the expected address" WARNING during `fly deploy`
- Switch `proxy_pass` to use a variable with `resolver 100.100.100.100 valid=30s` so nginx can start without resolving MagicDNS names at config load time
- DNS results cached 30s per worker — no per-request lookup overhead
## Context
The WARNING was a race condition: Fly checks for listeners right after the machine starts, but `start.sh` ran ~5-10s of Tailscale setup before starting nginx. The health check always passed later, but the warning was noisy.
## Test plan
- [ ] Merge and let the deploy-fly workflow trigger
- [ ] Check runner logs for absence of the WARNING
- [ ] Verify `docs.eblu.me` still serves correctly
- [ ] Verify `/healthz` still passes
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/128
## Summary
- Fix Immich Ingress `host: photos` causing 404 with ProxyGroup (same FQDN mismatch as Prometheus/Loki)
- Migrate Homepage from old per-service Tailscale proxy to shared ProxyGroup (was the last holdout)
- Add Immich and Navidrome to `services-check` HTTP endpoints
## Deployment Notes
- Already tested on branch: Immich and Homepage both return 200 via Caddy
- Homepage's old Helm-managed Ingress was deleted manually; ArgoCD may recreate it on sync — prune with `argocd app sync homepage --prune` after merge
- Old per-service `ts-homepage-*` pod in tailscale namespace can be cleaned up after confirming ProxyGroup works
## Test Plan
- [x] `curl https://photos.ops.eblu.me/` returns 200
- [x] `curl https://go.ops.eblu.me/` returns 200
- [ ] `mise run services-check` fully passes after merge
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/127
## Summary
- Introduce `tag:flyio-target` so services must explicitly opt in to be reachable by the fly.io proxy
- Replace broad `tag:k8s` and `tag:homelab` grants with the new tag in the ACL rule and test
- Add `tailscale.com/tags: "tag:k8s,tag:flyio-target"` annotation to docs, loki, and prometheus Ingresses
- Switch Alloy push endpoints from `*.ops.eblu.me` (Caddy) to `*.tail8d86e.ts.net` (Tailscale Ingress)
- Update docs: flyio-proxy, caddy, tailscale, forgejo (future public access + security checklist), expose-service-publicly
## Manual step (not in PR)
Update the k8s operator OAuth client in the Tailscale admin console to include `tag:flyio-target` in its scope. Without this, the operator cannot assign the new tag to Ingress proxy nodes.
## Deployment order
1. **Pulumi ACLs** — `mise run tailnet-preview && mise run tailnet-up`
2. **OAuth client** — Manual update in Tailscale admin console
3. **K8s Ingresses** — `argocd app sync apps && argocd app sync docs loki prometheus`
4. **Fly.io proxy** — `mise run fly-deploy`
5. **Verify** — `mise run services-check`, check Grafana dashboards
## Test plan
- [ ] `mise run tailnet-preview` shows clean diff
- [ ] `argocd app diff docs`, `argocd app diff loki`, `argocd app diff prometheus` show only annotation additions
- [ ] After deploy: Grafana dashboards show continued log/metric flow
- [ ] `curl -sf https://docs.eblu.me` returns 200
- [ ] `mise run services-check` passes
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/126
## Summary
- Convert all 12 `op item get ... --fields ... --reveal` calls in Ansible to the newer `op read "op://vault/item/field"` syntax
- Remove the `regex_replace` workaround on the Fly deploy token (no longer needed since `op read` returns clean unquoted values)
- Covers `ansible/playbooks/indri.yml`, `ansible/roles/caddy/tasks/main.yml`, `ansible/roles/jellyfin_metrics/tasks/main.yml`, and `ansible/roles/alloy/tasks/main.yml`
## Test plan
- [x] `mise run provision-indri -- --check --diff` dry run passes (ok=67, failed=0)
- [x] No `op item get` calls remain in `ansible/` directory
- [x] All pre-commit hooks pass (yaml, ansible-lint, TruffleHog, etc.)
- [ ] Full provision run after merge to confirm secrets resolve correctly
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/125
## Summary
- Tasks with a due date are now only shown when due today or earlier
- Recurring tasks stay hidden until their next occurrence is actionable
- Tasks without a due date continue to always display
## Test plan
- [x] Ran `mise run blumeops-tasks` — verified 18 undated tasks display correctly
- [x] Confirmed "BlumeOps doc review" (due tomorrow) is correctly hidden
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/124
## Summary
- Add public docs (`docs.eblu.me`) and Fly.io healthz checks to `mise run services-check`
## Test plan
- [ ] `mise run services-check` includes new "Public services (via Fly.io)" section
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/122
## Summary
- Strip literal quotes from Fly.io deploy token when syncing to Forgejo Actions secrets
- The `op` CLI wraps values containing spaces in quotes; the Fly.io token format is `FlyV1 <key>` (contains a space)
- This caused the CI deploy workflow to fail with a 401 auth error
## Test plan
- [x] `mise run provision-indri -- --tags forgejo_actions_secrets` succeeds
- [ ] Re-trigger deploy-fly workflow after merge — should authenticate successfully
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/121
## Summary
- Adds a Fly.io reverse proxy (`blumeops-proxy`) that tunnels public traffic to homelab services over Tailscale
- First service exposed: `docs.eblu.me` — the Quartz static docs site
- Includes Pulumi IaC for Tailscale auth key/ACLs and Gandi DNS CNAME
- Adds mise tasks (`fly-deploy`, `fly-setup`, `fly-shutoff`) and Forgejo CI workflow
## Key details
- Fly.io Firecracker VMs support TUN devices natively — no userspace networking needed
- Tailscale auth key is `preauthorized=True` to avoid device approval hangs on container restarts
- nginx caches aggressively for the static site; health check is on the default_server block
- ACLs restrict `tag:flyio-proxy` to `tag:k8s` on port 443 only
- DNS CNAME deployed and verified: `docs.eblu.me` → `blumeops-proxy.fly.dev`
## Test plan
- [x] `curl -sf https://blumeops-proxy.fly.dev/healthz` returns `ok`
- [x] `curl -I -H "Host: docs.eblu.me" https://blumeops-proxy.fly.dev/` returns 200 with `X-Cache-Status`
- [x] `curl -I https://docs.eblu.me/` returns 200 with valid Let's Encrypt cert
- [x] `dig forge.ops.eblu.me` still resolves to 100.98.163.89 (private services unaffected)
- [x] Set `FLY_DEPLOY_TOKEN` Forgejo Actions secret for CI auto-deploy
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/120
## Summary
- Adds `docs/how-to/expose-service-publicly.md` documenting the full plan for exposing `docs.eblu.me` to the public internet
- Covers Cloudflare Tunnel + CDN architecture, DNS migration from Gandi, Caddy TLS changes, Pulumi IaC, k8s cloudflared deployment, and verification steps
- Pattern is reusable for future public services
- Marked as "Plan — not yet implemented" status
## Test plan
- [x] `docs-check-links` passes
- [x] `docs-check-index` passes
- [x] All pre-commit hooks pass
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/118
## Summary
- Updated frontmatter `title:` in all 63 doc cards from slug-case to human-readable (e.g. `borgmatic` → `Borgmatic`, `ai-assistance-guide` → `AI Assistance Guide`)
- Titles now closely match file stems so `[[wiki-links]]` render naturally without alternate anchor text
- Corrected titles that diverged from stems (e.g. `host-inventory` → `Hosts`, `grafana-alloy` → `Alloy`, `argocd-applications` → `Apps`)
- Deleted `title-test-alpha.md` and `title-test-beta.md` test cards and removed their reference index entry
## Deployment and Testing
- [x] `docs-check-links` passes — all wiki-links valid
- [x] `docs-check-index` passes
- [x] `docs-check-filenames` passes
- [ ] Verify titles render correctly on docs site after deploy
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/117
## Summary
- Remove `docs-check-titles` pre-commit hook and mise task — wiki-links resolve by filename stem, not frontmatter title, so slug-format titles and uniqueness aren't needed
- Add two test cards (`title-test-alpha`, `title-test-beta`) with identical `title: Title Test Card` to verify duplicate titles don't break Quartz or obsidian.nvim
- Retitle `index.md` from `blumeops-documentation` to `BlumeOps`
- Add GitHub and Forgejo repo links to homepage intro
## Test plan
- [ ] Deploy to docs site and verify both test cards render and cross-link correctly
- [ ] Verify homepage title renders as "BlumeOps"
- [ ] Verify repo links on homepage work
- [ ] After confirming, remove test cards in a follow-up
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/116
## Summary
- New reference card (`docs/reference/infrastructure/gandi.md`) covering DNS records, Pulumi config, TLS integration
- New how-to guide (`docs/how-to/gandi-operations.md`) for DNS deployment and PAT cycling with `pbpaste` shortcut
- Rewritten homepage intro for wider audience ahead of public docs.eblu.me
- Cross-linked from reference index, routing, caddy, and how-to index
- Fixed PAT expiration inaccuracy in `pulumi/gandi/README.md` (max is 90 days, not 30)
## Test plan
- [ ] Verify wiki-links resolve in Quartz build
- [ ] Review gandi reference card for accuracy
- [ ] Review gandi-operations how-to for accuracy
- [ ] Check homepage reads well for external visitors
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/115
## Summary
- Add missing `ai` changelog fragment type to the update-documentation how-to guide (comment and table)
- Consolidate `cicd` → `ci-cd` tag on forgejo.md
- Consolidate `network` → `networking` tag on routing.md and tailscale.md
Found during random doc review via `docs-review-tags`.
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/114
## Summary
- Reviewed and cleaned up exploring-the-docs tutorial: simplified wiki-links, fixed broken replication/ reference, added Related section, corrected zk-docs flags to match CLAUDE.md
- Added orphan detection to doc-links (finds docs not linked from any other doc)
- Added new doc tooling: `doc-index` (checks category index coverage), `doc-stale` (staleness report), `doc-tags` (tag inventory)
- Added `doc-index` as a pre-commit hook
- Updated use-pypi-proxy to document env-var-based proxy toggle for pip/uv
- Updated ai-assistance-guide with new doc task descriptions
## Test plan
- [ ] Run `mise run doc-links` — passes
- [ ] Run `mise run doc-index` — passes
- [ ] Run `mise run doc-stale` — informational output
- [ ] Run `mise run doc-tags` — informational output
- [ ] Pre-commit hooks pass
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/112
## Summary
- Rename section index files to match their titles (tutorials.md, reference.md, how-to.md, explanation.md) so all filenames are unique
- Convert all ~47 path-based wiki-links to simple filename format across 15 files
- Update doc-filenames task to no longer skip index.md files
- Update doc-links task to reject path-based links containing '/'
This ensures all wiki-links work correctly in obsidian.nvim by making links resolvable by filename alone.
## Testing
- `mise run doc-filenames` - all unique
- `mise run doc-links` - no broken or path-based links
- `mise run doc-titles` - no duplicates
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/109
## Summary
- Add `docs/how-to/restart-indri.md` with safe shutdown and startup procedures
- Add `docs/reference/services/automounter.md` documenting the SMB share automounter app
- Update indri reference card with GUI applications section
- Update how-to and reference indexes
## Test plan
- [ ] Review restart-indri.md for accuracy
- [ ] Verify AutoMounter details are correct
- [ ] Perform the actual restart using this guide
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/108
## Summary
- New `forgejo_actions_secrets` Ansible role syncs repository-level Actions secrets from 1Password to Forgejo via the Forgejo API
- Replaces manual process of copying secrets from 1Password to Forgejo UI
- Documents the one-time PAT setup requirement in forgejo.md
## Manual Setup Required
Before this role can run, a Forgejo PAT must be created:
1. Go to https://forge.ops.eblu.me/user/settings/applications
2. Create a new token with `write:repository` scope
3. Store it in 1Password → "Forgejo Secrets" item → `api-token` field
This has already been done.
## Test Plan
- [x] Ran `mise run provision-indri -- --tags forgejo_actions_secrets` successfully
- [x] Verified secret synced (API returned 204 = updated existing)
- [x] Ansible-lint passes
🤖 Generated with [Claude Code](https://claude.ai/code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/107
## Summary
Fixes the "isn't yet tracked by git, dates will be inaccurate" warnings by using Quartz's `-d docs` flag instead of symlinking.
## Problem
The previous approach symlinked `content -> docs`, but git doesn't follow symlinks. When Quartz asked git about `content/index.md`, git had no history for that path.
## Solution
Use `npx quartz build -d docs` to tell Quartz to read from `docs/` directly. Now when Quartz asks git about `docs/index.md`, git finds the actual file history.
- CHANGELOG.md is copied (not symlinked) into `docs/` for the build, then removed
- All other files have accurate git-based dates
## Testing
Tested locally - build produces no warnings.
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/106
## Summary
Fixes the "isn't yet tracked by git, dates will be inaccurate" warnings in the Build docs step by restructuring how Quartz builds the documentation.
## Problem
Previously, we copied docs into Quartz's content folder. Since this was inside a fresh Quartz clone with no history of our files, the `CreatedModifiedDate` plugin couldn't determine accurate dates.
## Solution
Build Quartz from within the blumeops repo instead:
1. Copy Quartz's build system (quartz/, package.json, etc.) into the workspace
2. Symlink `content` -> `docs` (preserves git history)
3. Symlink `docs/CHANGELOG.md` -> `../CHANGELOG.md`
4. Build from workspace root where git can trace file history
5. Clean up artifacts after creating tarball
## Deployment and Testing
- [ ] Run build workflow and verify no "not tracked by git" warnings
- [ ] Verify file dates appear correctly on built docs site
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/105
## Summary
- Add `version_type` choice input with options: BUMP_PATCH (default), BUMP_MINOR, BUMP_MAJOR, SPECIFIC_VERSION
- Add optional `specific_version` input for explicit version selection
- Include changelog content in Forgejo release body under "What's Changed" section
- Move CHANGELOG.md to repository root (still copied into docs during Quartz build)
- Add CHANGELOG link to docs index page
- Update doc-links script to recognize build-time docs from repo root
## Changes
**Workflow inputs:**
- Previously: single optional `version` string input
- Now: `version_type` choice dropdown (defaults to BUMP_PATCH) + optional `specific_version` for explicit versions
**Release body:**
- Previously: just asset download instructions
- Now: includes "What's Changed" section with changelog entries for this release
**CHANGELOG.md location:**
- Previously: `docs/CHANGELOG.md`
- Now: `CHANGELOG.md` (repo root), copied into docs content during build
## Deployment and Testing
- [ ] Run build workflow with BUMP_PATCH (default)
- [ ] Run build workflow with BUMP_MINOR
- [ ] Verify changelog appears in release body
- [ ] Verify docs site includes CHANGELOG page
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/104
## Summary
- Rename `indri-services-check` task to `services-check` since it checks all services (indri native, Kubernetes, HTTP endpoints), not just indri-specific ones
- Update references in CLAUDE.md, ai-assistance-guide.md, and troubleshooting.md
## Deployment and Testing
- [ ] Run `mise run services-check` to verify the task works under its new name
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/103
## Summary
- Add "Forgejo Actions Secrets" section to forgejo reference card
- Document that `ARGOCD_AUTH_TOKEN` is used by `build-blumeops.yaml` workflow
- Note that secrets are stored in 1Password but manually copied to Forgejo (no auto-sync)
- Add missing `build-blumeops.yaml` to workflows list
- Clarify distinction between server config secrets (1Password → Ansible) vs CI/CD secrets (Forgejo UI)
## Context
The forgejo-runner ArgoCD app was showing OutOfSync because a previous attempt stored `argocd_token` in the ExternalSecret. This was incorrect - the token is actually a Forgejo Actions secret, not a k8s secret. Synced the app to remove the drift and added documentation to prevent future confusion.
🤖 Generated with [Claude Code](https://claude.ai/code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/102
## Summary
- Fix env var name from `ND_SCANSCHEDULE` to `ND_SCANNER_SCHEDULE` (Navidrome uses viper config where dots become underscores)
- Use explicit `@every 1h` format for clarity
- Reorder CLAUDE.md rules to emphasize running zk-docs first
## Root Cause
Navidrome logs showed "Periodic scan is DISABLED" at startup despite the env var being set. The config key is `scanner.schedule`, which translates to `ND_SCANNER_SCHEDULE` (not `ND_SCANSCHEDULE`).
## Deployment and Testing
- [ ] Sync navidrome app: `argocd app sync navidrome`
- [ ] Verify pod restarts with new env var
- [ ] Check logs for "Scheduling scanner" message instead of "Periodic scan is DISABLED"
- [ ] Wait ~1 hour and confirm scan runs automatically
🤖 Generated with [Claude Code](https://claude.ai/code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/101
## Summary
- Remove ~/Pictures from borgmatic source directories
- Update borgmatic and backup policy documentation
- Add Sifaka-Native Data section to clarify that photos (via Immich), music (via Navidrome), and video (via Jellyfin) are stored directly on Sifaka
## Deployment and Testing
- [ ] Run `mise run provision-indri -- --tags borgmatic --check --diff` to preview changes
- [ ] Run `mise run provision-indri -- --tags borgmatic` to apply
- [ ] Verify borgmatic config no longer includes ~/Pictures
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/100
## Summary
- Add the troubleshooting how-to guide to the zk-docs script output
- Placed after how-to/index.md to group how-to content together
## Deployment and Testing
- [x] Verified script runs successfully with `mise run zk-docs`
- [x] Changelog fragment added
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/99
## Summary
- Add `doc-random` mise task that selects a random documentation card for review
- Add how-to/knowledgebase section with review-documentation guide
- Add Caddy reference card with proxy configuration details
- Fix replication tutorial sequence (tailscale-setup now links to core-services)
- Fix "BluemeOps" typo in tailscale-setup
- Clean up obsolete zk/ directory references from doc-links
## Deployment and Testing
- [x] `mise run doc-random` works and displays a random card
- [x] `mise run doc-links` passes (all wiki-links valid)
- [x] Pre-commit hooks pass
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/98
## Summary
- Delete `docs/zk/` directory - all useful content migrated to structured docs
- Delete `docs/README.md` - `docs/index.md` is now the documentation root
- Add `devpi` reference card and `use-pypi-proxy` how-to guide
- Add maintenance notes to `indri` reference (sleep prevention, passwordless sudo)
- Add iCloud Photos backup note to `borgmatic` reference
- Rewrite `zk-docs` mise task to prime AI context with key docs instead of legacy cards
- Update `CLAUDE.md` and `README.md` to remove zk references
- Update `exploring-the-docs` with AI context priming section
This completes the Diataxis documentation restructuring. All six phases are now done.
## Deployment and Testing
- [x] Pre-commit hooks pass (including doc-links validator)
- [ ] Build and deploy to docs.ops.eblu.me to verify rendering
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/97
## Summary
- Create `docs/explanation/` directory with index and three explanation articles
- why-gitops: Philosophy of GitOps for homelabs (memory, rollback, AI context)
- architecture: How pieces fit together (ASCII diagrams of hosts, data flow, secrets)
- security-model: Tailscale zero-trust, 1Password secrets, access control philosophy
- Update docs/index.md with How-to and Explanation section links
- Update exploring-the-docs to link Explanation section
Decision log deferred to future work.
## Deployment and Testing
- [x] Pre-commit hooks pass (including doc-links validator)
- [ ] Build and deploy to docs.ops.eblu.me to verify rendering
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/96
## Summary
- Create `docs/how-to/` directory with index and four how-to guides
- deploy-k8s-service: Quick reference for Kubernetes deployments via ArgoCD
- add-ansible-role: Adding new Ansible roles for indri services
- update-tailscale-acls: Modifying Tailscale ACL policies via Pulumi
- troubleshooting: Diagnosing and fixing common issues
- Update exploring-the-docs to include How-to section links
- Update README.md to mark Phase 4 as complete
## Deployment and Testing
- [x] Pre-commit hooks pass (including doc-links validator)
- [ ] Build and deploy to docs.ops.eblu.me to verify rendering
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/95
The towncrier config uses the type's `directory` field as the type
identifier in filenames, NOT as subdirectories. Correct format:
docs/changelog.d/<name>.<type>.md
NOT:
docs/changelog.d/<type>/<name>.md
- Move fragments to root with type suffix
- Remove empty type subdirectories
- Fix CLAUDE.md instructions
- Fix tutorial examples in contributing.md and ai-assistance-guide.md
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Fragments in subdirectories should be named `<name>.md`, not
`<name>.<type>.md` - the type is already indicated by the directory.
- Rename feature/auto-deploy-docs.feature.md → feature/auto-deploy-docs.md
- Rename misc/+container-tag-no-confirm.misc.md → misc/+container-tag-no-confirm.md
- Update CLAUDE.md with correct fragment path format
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This ensures ArgoCD sync triggers a pod rollout when the URL changes,
since ConfigMap data changes don't restart pods automatically.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add -L flag to follow redirects
- Add -f flag to fail on HTTP errors
- Use dpkg --print-architecture as fallback for TARGETARCH
- Verify binary works after download
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
## Summary
- Add `uv` and `argocd` CLI to forgejo-runner container image
- Add `workflow-bot` ArgoCD account with sync permissions (declarative via kustomize patches)
- Add `ARGOCD_AUTH_TOKEN` to forgejo-runner external secret for workflow auth
- Update build workflow to auto-deploy docs after release:
- Update configmap with new release URL
- Commit changelog and configmap changes
- Sync docs app via ArgoCD
## Deployment and Testing
Manual steps required before this can work:
1. [ ] Build and push new forgejo-runner image (v2.4.0)
2. [ ] Sync argocd app to create workflow-bot account
3. [ ] Generate token: `argocd account generate-token --account workflow-bot`
4. [ ] Store token in 1Password under "Forgejo Secrets" with field `argocd_token`
5. [ ] Sync forgejo-runner app to pick up new external secret
6. [ ] Update forgejo-runner deployment to use new image version
7. [ ] Test by running workflow manually
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/93
- Use [[reference/index | reference]] for ambiguous index.md link
- Update doc-links to validate both filenames and paths
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Convert all wiki-links from title-based to filename-based
- Update doc-links to validate against filenames
- Add doc-filenames task for duplicate filename detection
- Consolidate doc hooks into single local block in pre-commit config
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Wiki-links with display text must use spaces around the pipe:
- Good: [[target | display]]
- Bad: [[target|display]]
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
## Summary
- Convert all frontmatter titles to lowercase-hyphenated format (e.g., `grafana-alloy` instead of `Grafana Alloy`)
- Update all wiki-links to use the new slug format
- Update `doc-titles` task to validate slug format (lowercase, hyphens only)
Quartz appears to require titles without spaces for wiki-link resolution.
## Deployment and Testing
- [x] Pre-commit hooks pass (`doc-titles` and `doc-links`)
- [ ] Build docs v1.0.8 and deploy
- [ ] Verify wiki-links resolve correctly (e.g., `[[grafana-alloy]]`)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/92
## Summary
- Remove aliases from all zk cards to prevent them from capturing wiki-links
- Convert all wiki-links from `[[filename|Title]]` to `[[Title]]` format
- Replace `doc-filenames` task with `doc-titles` for duplicate title detection
- Update pre-commit hook to use `doc-titles`
Wiki-links now resolve to reference docs by their frontmatter title, which is more readable and maintainable than filename-based links.
## Deployment and Testing
- [x] Pre-commit hooks pass (including new `doc-titles` check)
- [x] Manually verified zk cards have aliases removed
- [ ] Deploy docs v1.0.7 and verify wiki-links resolve correctly
- [ ] Test links to reference docs (e.g., [[Grafana Alloy]], [[ArgoCD]])
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/91
## Summary
- Quartz's "shortest" path mode resolves wiki-links by **filename**, not frontmatter title
- Previous PR used title-based links like `[[Grafana Alloy]]` which looked for non-existent `Grafana-Alloy.md`
- Now using filename-based links like `[[alloy|Grafana Alloy]]` which correctly resolve
## Changes
- Rename zk duplicate files with `-log` suffix (e.g., `argocd.md` → `argocd-log.md`)
- Rename `reference/storage/postgresql.md` to `postgresql-storage.md`
- Convert all 175 wiki-links from `[[Title]]` to `[[filename|Title]]` format
- Rename `doc-card-titles` task to `doc-filenames` (checks filename uniqueness, not titles)
- Update pre-commit hook for renamed task
## Deployment and Testing
- [x] Pre-commit hooks pass
- [x] `mise run doc-filenames` shows no duplicate filenames
- [ ] Verify wiki-links work correctly in Quartz build
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/90
## Summary
- Add `doc-card-titles` mise task to enumerate all doc cards by title/id and detect duplicates
- Remove redundant aliases from zk cards where alias matched the id
- Rename `reference/storage/postgresql.md` title to "PostgreSQL Storage" to avoid duplicate with `reference/services/postgresql.md`
- Convert all 175 path-based wiki-links `[[reference/path|Title]]` to title-based `[[Title]]` format
- Add pre-commit hook to check for duplicate card titles on doc changes
## Deployment and Testing
- [x] Pre-commit hooks pass
- [x] `mise run doc-card-titles` shows no duplicates
- [ ] Verify wiki-links work correctly in Quartz build
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/89
Links like [[services/foo]] need to be [[reference/services/foo]]
to resolve correctly from the docs root.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add navigation to Reference section and clarify that README is
temporary home during Diataxis restructuring.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
## Summary
- Move Reference from Phase 4 to Phase 2 in the documentation restructuring plan
- Update directory layout in README to reflect new phase order
- Add explanatory note about why Reference comes first (so other docs can link to it)
## Deployment and Testing
- [x] Updated docs/README.md with new phase order
- [x] Added towncrier changelog fragment
- [ ] Ready to proceed with Phase 2 (Reference) content after merge
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/87
## Summary
- Configure towncrier with custom types (feature, bugfix, infra, doc, misc)
- Build initial v0.1.0 changelog from zk management log entries
- Integrate towncrier into build-blumeops workflow
- Update README to mark Phase 1b complete
## How It Works
1. Add changelog fragments to `docs/changelog.d/` as `<id>.<type>.md`
2. When running build-blumeops workflow, towncrier collects fragments
3. CHANGELOG.md is updated and fragments are removed
4. Changes are committed back to main before docs build
## Testing
- [x] Tested `uvx towncrier build` locally
- [ ] Test workflow execution (after merge)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/86
Mark Phase 1a as complete with link to first release.
Move CHANGELOG to Phase 1b with towncrier note.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add Authorization header using GITHUB_TOKEN
- Remove silent fail flag to see error responses
- Log API responses for debugging
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Remove -f flag from curl so 404 on /releases/latest doesn't fail the
script when there are no releases yet.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
## Summary
- Move all existing zettelkasten cards from `docs/` to `docs/zk/` as a temporary holding area
- Update `zk-docs` mise task to look in the new location
- Add `docs/README.md` explaining the Diataxis-based restructuring plan and target audiences
## Context
This is phase 1 of a multi-phase documentation restructuring effort. The goal is to reorganize docs to follow the Diataxis framework while serving multiple audiences:
1. Erich (owner) - knowledge graph/zk
2. Claude/AI agents - memory and context enrichment
3. New external readers - high-level overview
4. Potential operators/contributors - onboarding
5. Replicators - people wanting to duplicate the approach
## Testing
- [x] Verified `mise run zk-docs` still works with the new path
- [x] Updated obsidian.nvim config (in ~/.config/nvim) to point to new path
## Note
The obsidian.nvim config change is outside this repo but was made as part of this work.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/84
## Summary
- Add "Unhealthy Pods" stat panel showing count of pods in error states (ImagePullBackOff, CrashLoopBackOff, etc.) with red background when > 0
- Add "Pods by Waiting Reason" time series chart showing container waiting states over time
- Provides visibility into stuck pods that ArgoCD doesn't track (since it manages CronJobs, not the Jobs/Pods they spawn)
## Context
This addresses the issue where a `zim-watcher` cronjob pod was stuck in `ImagePullBackOff` for 11 days without any alerting. ArgoCD showed the CronJob as "Synced, Healthy" because it only manages the CronJob resource, not its spawned Jobs/Pods.
## Deployment and Testing
- [ ] Sync grafana-config app to test branch
- [ ] Verify dashboard renders correctly
- [ ] Confirm "Unhealthy Pods" shows 0 (green) when no issues
- [ ] Reset to main after merge
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/83
## Summary
- Move 21 blumeops-tagged zettelkasten cards from ~/code/personal/zk/ to docs/
- Create symlink ~/code/personal/zk/blumeops -> blumeops/docs for obsidian integration
- Update zk-docs mise task to read from local docs/ directory
- Add blumeops workspace to obsidian.nvim config (strict=true)
## Benefits
- Docs are now git-managed in the blumeops repo (visible on GitHub)
- Wiki links between blumeops docs continue to work via symlink
- obsidian-sync isolation: docs don't sync to work laptop
- Direct editing via obsidian.nvim with dedicated workspace
## Testing
- [x] Files moved to docs/ (21 files)
- [x] Symlink created: ~/code/personal/zk/blumeops -> blumeops/docs
- [x] zk-docs mise task updated and working
- [ ] Verify obsidian.nvim link resolution (after merge)
- [ ] Verify obsidian backlinks work
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/82
Use Quick Launch settings for Kagi search with suggestions instead of
the search widget, which is the proper way to configure keyboard-driven
search in homepage.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
## Summary
- Deploy Navidrome music streaming server to k8s
- NFS mount for music library from sifaka:/volume1/music (read-only)
- Local PVC for SQLite database and config (10Gi)
- Tailscale ingress for dj.tail8d86e.ts.net
- Caddy reverse proxy for dj.ops.eblu.me
- Homepage annotations for dashboard discovery in Media group
## Deployment and Testing
- [ ] Sync `apps` application to pick up new Application definition
- [ ] Set navidrome app to feature branch and sync
- [ ] Verify NFS mount with `kubectl exec`
- [ ] Provision Caddy for dj.ops.eblu.me
- [ ] Access https://dj.ops.eblu.me and create initial admin user
- [ ] Verify Homepage shows DJ in Media group
- [ ] Reset to main and resync after merge
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/79
## Summary
- Fix ArgoCD icon (use `argo-cd.png` per Dashboard Icons naming)
- Add Borgmatic backup metrics widget (time since last backup, archive size)
- Add Sifaka NAS disk usage widget (used/total space)
- Create `[[grafana]]` zk card with management notes
## What didn't work
Attempted Grafana iframe embedding for a metrics panel but reverted:
- Homepage iframe widget only supports height classes, not width
- Some panels fail to load even with anonymous auth enabled
- Documented in grafana zk card for future reference
## Deployment and Testing
- [x] ArgoCD icon displays correctly
- [x] Borgmatic metrics show time since backup and archive size
- [x] NAS disk usage shows used/total bytes
- [x] Grafana reverted to authenticated-only access
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/76
## Summary
- Remove hajimari (unmaintained since Oct 2022, broken helm deps)
- Add gethomepage (28k stars, actively maintained, monthly releases)
- Migrate custom apps, bookmarks, and search config
- Enable k8s RBAC for service autodiscovery
- Configure Tailscale ingress at go.tail8d86e.ts.net
## Why the switch
Hajimari hasn't released since October 2022. The helm chart has a broken
dependency (bjw-s/common URL is 404), and unreleased code on main has bugs.
gethomepage has similar k8s autodiscovery via ingress annotations and is
very actively maintained.
## Deployment and Testing
- [ ] Delete hajimari app from ArgoCD
- [ ] Delete hajimari namespace
- [ ] Sync apps to pick up new homepage app
- [ ] Sync homepage app
- [ ] Verify go.ops.eblu.me loads
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/75
- Use chart from forge.ops.eblu.me/eblume/hajimari fork
- Use custom image from registry.ops.eblu.me/blumeops/hajimari
- Enables future customizations (search auto-focus, weather widget)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Pre-install skopeo for pushing images to zot registry.
Docker 27's manifest format has compatibility issues with zot,
so we use skopeo for the push step.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
## Summary
- Remove Tailscale sidecar from build-push-image action - registry.ops.eblu.me is directly reachable from k8s pods via Caddy
- Use skopeo for pushing images instead of docker push - Docker 27's manifest format has compatibility issues with zot registry
- Remove tailscale_authkey secret requirement from workflows
## Deployment and Testing
- [x] Tested with nettest-v0.10.0 tag - build succeeded and image pushed to registry
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/74
Avoids duplicate "Infrastructure" groups since Hajimari doesn't
merge customApps with discovered apps.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Set Kagi as default search provider with simple-icons:kagi
- Add Google and DuckDuckGo as alternative search providers
- Explicitly enable showAppGroups, showAppUrls, showAppInfo
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
## Summary
- Fix immich app to track `main` branch instead of `feature/immich` for values
- The tailscale-operator ignoreDifferences schema drift will be fixed by syncing the `apps` app
## Deployment and Testing
- [ ] Sync `apps` to fix tailscale-operator schema drift
- [ ] Sync `immich` to pick up correct image versions from main
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/71
## Summary
- Update all metrics role defaults to install scripts to ~/.local/bin following XDG conventions
- Scripts already manually moved on indri from ~/bin to ~/.local/bin
- Cleaned up orphaned scripts (devpi-metrics, transmission-metrics, mcquack) and plist files
## Deployment and Testing
- [x] Manually moved scripts on indri
- [x] Deleted orphaned plist files (devpi-metrics, devpi, kiwix-serve, transmission-metrics)
- [x] Deleted orphaned scripts (devpi-metrics, transmission-metrics, mcquack)
- [x] Verified no metrics dependencies on orphaned scripts (checked alloy config and textfile directory)
- [ ] Run ansible to update LaunchAgent plist files with new paths
- [ ] Verify metrics collection continues working
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/70
## Summary
- Add fixed Y-axis (0-220M) so the 200M autovacuum threshold is always visible
- Add dashed threshold lines at 150M (yellow warning) and 200M (red danger)
- Update title to clarify the threshold
## Context
The raw XID age naturally trends upward between vacuum freezes, which looked alarming without context. Current values (~143K-216K) are at 0.1% of the threshold - completely healthy.
## Deployment and Testing
- [ ] Sync grafana-config app to feature branch
- [ ] Verify threshold lines appear on PostgreSQL dashboard
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/69
1Password Connect doesn't support ?ssh-format=openssh, so we need a
separate Secure Note item with the OpenSSH-formatted key.
Created new 1Password item: argocd-forge-ssh-key
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Delete 13 .yaml.tpl files replaced by ExternalSecrets
- Update immich/README.md with direct CNPG secret copy instructions
- Update miniflux/README.md with context flag and ESO note
Only 1password-connect/secret-credentials.yaml.tpl remains (bootstrap).
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
## Summary
- Upgrades Immich image tag from v2.4.1 to v2.5.0
## Deployment and Testing
- [ ] Point immich ArgoCD app at feature branch and sync
- [ ] Verify pods come up healthy
- [ ] Verify Immich web UI accessible
- [ ] Reset to main and sync after merge
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/64
## Summary
- Update all references from `registry.tail8d86e.ts.net` to `registry.ops.eblu.me`
- Remove `tailscale_serve` ansible role (no longer needed - all services migrated to Caddy)
- Update minikube containerd config for new registry URL
- Update devpi manifest, CI actions, and mise tasks
## Deployment and Testing
- [ ] Run `mise run provision-indri -- --check --diff` (dry run)
- [ ] Run `mise run provision-indri -- --tags minikube` to update containerd config
- [ ] Sync devpi ArgoCD app: `argocd app sync devpi`
- [ ] Manually remove old Tailscale serve entry: `ssh indri 'tailscale serve --service=svc:registry off'`
- [ ] Test registry access: `curl https://registry.ops.eblu.me/v2/_catalog`
- [ ] Run `mise run indri-services-check` to verify all services healthy
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/58
## Summary
- Update CLAUDE.md with new service routing documentation
- Document the two DNS domains: `*.ops.eblu.me` (Caddy) vs `*.tail8d86e.ts.net` (Tailscale)
- Fix incorrect service listings (Prometheus/Loki are in k8s, not indri)
## ZK Updates (not in this PR)
Also updated the blumeops zk card with:
- Source code URL (forge is primary, GitHub is mirror)
- Services split into Caddy vs Tailscale sections
- Updated port map for Caddy
- Updated "Adding a New Service" instructions
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/57
## Summary
- Add layer4 TCP proxy configuration to Caddyfile template for SSH services
- Configure Forgejo SSH on port 2222 → localhost:2200
- Switch HTTPS from port 8443 (testing) to 443 (production)
- Requires Caddy rebuilt with `github.com/mholt/caddy-l4` plugin
## What This Enables
Git+SSH access via `forge.ops.eblu.me:2222` is now accessible from:
- Tailnet clients (gilbert)
- Docker containers on indri
- Kubernetes pods in minikube
This solves the DNS resolution issues where containers couldn't reach Tailscale MagicDNS names.
## Testing Done
- [x] Caddy rebuilt with layer4 plugin
- [x] Validated Caddyfile syntax
- [x] Cleared `svc:forge` from tailscale serve
- [x] Verified HTTPS works: `curl https://forge.ops.eblu.me`
- [x] Verified SSH works: `ssh -p 2222 forgejo@forge.ops.eblu.me`
- [x] Verified git clone works via new endpoint
- [x] Verified minikube pods can reach both HTTPS and SSH endpoints
## Deployment
Caddy is already running with the new config on indri. This PR captures the ansible changes.
## Next Steps
- Update zk docs with new git remote format
- Migrate registry and other services to Caddy
- Retire tailscale_services ansible role
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.tail8d86e.ts.net/eblume/blumeops/pulls/56
## Summary
- Add Caddy ansible role following zot pattern (manual build, ansible deploy)
- Caddy built with Gandi DNS plugin for ACME DNS-01 challenges
- Gandi PAT fetched from 1Password and written to secured file on indri
- Configure wildcard TLS for `*.ops.eblu.me`
- Initial services: forge, registry (indri-local)
- Uses port 8443 during testing to avoid Tailscale serve conflicts
## Build Instructions (already done)
On indri:
```bash
cd ~/code/3rd/caddy && mise run build
```
## Deployment and Testing
- [ ] Review Caddyfile configuration
- [ ] Run `mise run provision-indri -- --tags caddy` to deploy
- [ ] Test: `curl -v https://forge.ops.eblu.me:8443` (should get TLS cert)
- [ ] Test: `curl -v https://registry.ops.eblu.me:8443/v2/` (should return `{}`)
- [ ] Once verified, switch to port 443 and migrate services from Tailscale serve
## Files Changed
- `ansible/playbooks/indri.yml` - Add pre_task for Gandi PAT, add caddy role
- `ansible/roles/caddy/` - New role with Caddyfile and LaunchAgent templates
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.tail8d86e.ts.net/eblume/blumeops/pulls/55
## Summary
- Restructure Pulumi into separate projects: `pulumi/tailscale/` and `pulumi/gandi/`
- Add Gandi LiveDNS management for `eblu.me` domain
- Create wildcard DNS record `*.ops.eblu.me` → indri's Tailscale IP (100.98.163.89)
- Add mise tasks: `dns-up`, `dns-preview`
- Update `tailnet-up` to pass `--yes` by default
- Document PAT cycling process (expires every 30 days)
## Background
This enables using real DNS names (`*.ops.eblu.me`) that resolve to Tailscale IPs,
which allows containers and other systems to resolve services without depending on
MagicDNS. Since Tailscale IPs (100.x.x.x) are not publicly routable, services remain
tailnet-only while using standard DNS.
## Deployment and Testing
- [ ] Run `cd pulumi/gandi && uv sync` to install dependencies
- [ ] Run `cd pulumi/gandi && pulumi stack init eblu-me` to create stack
- [ ] Run `mise run dns-preview` to verify configuration
- [ ] Run `mise run dns-up` to apply DNS records
- [ ] Verify with `dig +short test.ops.eblu.me` returns `100.98.163.89`
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.tail8d86e.ts.net/eblume/blumeops/pulls/54
Use simple grep and awk to parse plain text tailscale status output
instead of trying to parse JSON. Also show the status output for
debugging.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Use 'tailscale status' to get indri's Tailscale IP and add it to
/etc/hosts for registry hostname resolution. The registry service
runs on indri, so we need indri's IP specifically.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
The Tailscale container's DNS doesn't work because it runs in userspace
mode. Instead, resolve the registry IP using 'tailscale ip' and add it
to /etc/hosts inside the container before running skopeo.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add tag:ci-gateway to tagOwners
- Grant ci-gateway access to registry on port 443
- Add test for ci-gateway -> registry access
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Docker Desktop's VM can't resolve tailnet hostnames. Work around this by:
1. Starting a Tailscale container that joins the tailnet
2. Building the image with docker build
3. Saving to tarball with docker save
4. Pushing via skopeo inside the Tailscale container
Uses TS_CI_GATEWAY_AUTHKEY repository secret for authentication.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Fetches logs for Forgejo Actions runs from indri's local storage.
Logs are stored as zstd-compressed files in the forgejo data directory.
Usage: mise run indri-runner-logs <run_id>
Only works for runs executed by the indri-host-runner.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
## Summary
- Add `containers/nettest/` with Alpine-based Dockerfile and connectivity test script
- Add `.forgejo/workflows/build-nettest.yaml` workflow triggered by `nettest-v*` tags
- Test script checks DNS resolution and HTTPS connectivity to forge and registry
## Deployment and Testing
- [ ] Merge PR to main
- [ ] Run `mise run container-release nettest v0.1.0` to trigger first build
- [ ] Verify workflow runs successfully and container can reach tailnet services
- [ ] Manually test from minikube: `kubectl run nettest --rm -it --image=registry.tail8d86e.ts.net/blumeops/nettest:v0.1.0`
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.tail8d86e.ts.net/eblume/blumeops/pulls/52
## Summary
- Replace Docker with Buildah for container image builds
- No Docker socket required - buildah is daemonless
- Cleaner security model (no privileged containers or socket mounting)
- Remove Docker-related security context from deployment
## Changes
- Update Dockerfile to install buildah/podman instead of docker-cli
- Configure buildah storage with overlay driver and fuse-overlayfs
- Update composite action to use `buildah bud` and `buildah push`
- Add `imagePullPolicy: Always` to ensure fresh image pulls
- Update test workflow to verify buildah/podman
## Testing
- [ ] Runner pod starts successfully
- [ ] Buildah is available in runner
- [ ] Test workflow verifies buildah/podman versions
- [ ] Container build workflow builds and pushes to zot
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Reviewed-on: https://forge.tail8d86e.ts.net/eblume/blumeops/pulls/51
2026-01-24 13:30:26 -08:00
784 changed files with 60742 additions and 36990 deletions
description: Classifies proposed changes as C0/C1/C2 before work begins. Use proactively when the user describes a new task or change, before any implementation starts.
tools: Read, Glob, Grep, Bash
model: haiku
permissionMode: dontAsk
---
You are a change classifier for the BlumeOps infrastructure project. Your job is to assess a proposed change and classify it as C0, C1, or C2 before any work begins.
## Classification Criteria
| Class | Name | When to use | Key trait |
|-------|------|-------------|-----------|
| **C0** | Quick Fix | Small, low-risk, fix-forward safe | Direct to main, no PR |
| **C1** | Human Review | Moderate complexity or risk | Feature branch + PR, docs-first |
description: Infrastructure health monitor. Use proactively after deployments, provisioning, or when the user asks about service status. Runs services-check and diagnoses failures.
tools: Bash, Read, Grep, Glob
model: haiku
permissionMode: dontAsk
background: true
---
You are an infrastructure health monitor for the BlumeOps homelab.
When invoked, run the full health check suite and report results:
1. Run `mise run services-check` and capture the full output
2. Parse the results — identify any FAILED services
3. For each failure, provide a brief diagnosis:
- Is the service process down?
- Is it a network/connectivity issue?
- Is it an ArgoCD sync issue?
4. Summarize: total services checked, how many passed, how many failed
If everything is healthy, keep the summary to one line.
If there are failures, group them by category:
- **Process failures** (service not running)
- **HTTP failures** (endpoint not responding)
- **Kubernetes failures** (pod not running, sync issues)
- **Connectivity failures** (SSH, network)
Do NOT attempt to fix anything. Report findings only.
Context:
- Services run across indri (Mac Mini, native + minikube), ringtail (NixOS, k3s), and Fly.io
- Use `--context=minikube-indri` for indri k8s commands, `--context=k3s-ringtail` for ringtail
- HTTP endpoints are proxied through Caddy at `*.ops.eblu.me`
- Public endpoints go through Fly.io at `*.eblu.me`
description: Mikado chain navigator for C2 changes. Use when resuming a C2 chain, checking chain status, or deciding which leaf node to work next. Understands the Mikado Branch Invariant.
tools: Read, Glob, Grep, Bash
model: sonnet
permissionMode: dontAsk
---
You are a Mikado chain navigator for the BlumeOps C2 change process. You help the user understand the current state of a Mikado chain and decide what to do next.
## What You Do
1. Run `mise run docs-mikado --resume` to detect the current chain state
2. Read the relevant Mikado cards (docs in `docs/how-to/` with `status: active`)
3. Analyze the dependency graph and branch position
4. Recommend the next action
## Chain State Analysis
After running `docs-mikado --resume`, interpret the output:
- **Planning phase:** Cards are being added, no code yet. Suggest reviewing the dependency graph for completeness.
- **Mid-cycle:** An `impl` is in progress. Identify which leaf is being worked and what remains.
- **Between cycles:** A leaf was just closed. Identify the next ready leaf and summarize what it requires.
- **Finalized:** The chain is complete and awaiting merge.
- **Invariant violation:** A plan commit was found after impl. Explain the reset procedure.
## Recommending Next Actions
For each ready leaf node:
1. Read the card content to understand what it requires
2. Check if there are related source files (manifests, playbooks, configs)
3. Assess relative complexity and suggest an ordering if multiple leaves are ready
4. Note any potential risks or dependencies not captured in the card graph
Guidance for AI agents working in this repository. See also [[ai-assistance-guide]].
## Overview
blumeops is Erich Blume's GitOps repository for personal infrastructure, orchestrated via tailnet `tail8d86e.ts.net`.
**CRITICAL: Public repo at github.com/eblume/blumeops - never commit secrets!**
**Shell:** The user's interactive shell may differ from the current harness shell. Prefer repo-safe, non-interactive commands when possible, and match the user's shell conventions when giving interactive examples.
## Rules
1. **Always run `mise run ai-docs` at session start**
This will refresh your context with important information you will be assumed to know and follow.
**Read the full output** — never truncate, pipe to `head`/`tail`, or skip sections.
For problems with a large surface area, ask the user if `mise run ai-sources` should also be run — it concatenates all non-doc source files (~270K tokens) for deep codebase context.
2. **Always use `--context=minikube-indri` with kubectl** (or `--context=k3s-ringtail` for ringtail services) - work contexts must never be touched
**NEVER run `minikube delete`** — it destroys all PVs, etcd, and cluster state. Use `minikube stop`/`minikube start` for restarts. If minikube is stuck, see [[restart-indri]]. Full rebuild from scratch requires the DR procedure in [[rebuild-minikube-cluster]].
3. **Classify the change as C0/C1/C2 before starting** (see below) — this determines branching and PR requirements
4. **Feature branches + PRs for C1/C2** - checkout main, pull, create branch, open PR via `tea pr create`. C0 goes direct to main.
5. **Check PR comments with `mise run pr-comments <pr_number>`** before proceeding
**C0** — commit directly to main. No branch or PR needed. Fix forward if problems arise.
**C1** — feature branch with early PR. Search related docs first, write documentation changes before code, deploy from the unmerged branch (ArgoCD `--revision`, Ansible from checkout). Upgrade to C2 if complexity spirals.
**C2** — branch `mikado/<chain-stem>` governed by the Mikado Branch Invariant: all card commits first, then code progress, then card closures. Commits use `C2(<chain>): plan/impl/close/finalize` convention. Reset the branch when new prerequisites are discovered. Resume with `mise run docs-mikado --resume`.
See [[agent-change-process]] for the full methodology.
## Project Structure
```
./docs/ # documentation (Diataxis, Quartz)
./docs/changelog.d/ # towncrier fragments
./.dagger/ # dagger pipelines
./.forgejo/ # forgejo-runner actions and workflows
Other code paths will be listed via ai-docs, this is just an overview. When you
encounter wiki-links (`[[like-this]]`) it is referring to docs/ cards.
## Service Deployment
### Kubernetes (ArgoCD)
Most services run in minikube on indri via ArgoCD (app-of-apps, manual sync). GPU workloads (Frigate, ntfy) run on ringtail's k3s cluster, also managed by ArgoCD.
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
## Project Overview
blumeops is Erich Blume's GitOps repository for personal infrastructure management, orchestrated via tailnet `tail8d86e.ts.net`.
**Critical: This repository is published publicly at https://github.com/eblume/blumeops, so never include any secrets!**
## Rules
1. At the start of every session, even if the user asked to do something else, run `mise run zk-docs -- --style=header --color=never --decorations=always` in order to review the `blumeops` documentation in the zettelkasten (zk). zk lives at `~/code/personal/zk`, and is managed via obsidian-sync (not git).
2. When making any changes, start by making sure you're on the `main` git branch and up-to-date, and then create a feature branch. Commit often while working, and create a PR using:
```fish
tea pr create --title "Description of change" --description "$(cat <<'EOF'
## Summary
- First change
- Second change
## Deployment and Testing
- [x] Done thing one
- [ ] Needed thing two
🤖 Generated with [Claude Code](https://claude.com/claude-code)
EOF
)"
```
The user will review your work as you go, and will merge the PR as the last step in the process, even after deploying. After the user reviews the PR and leaves comments, check for unresolved comments with:
```fish
mise run pr-comments <pr_number>
```
Address each unresolved comment before proceeding. The user will resolve comments on the Forge UI as they are addressed.
3. Always keep the zk cards up to date with any changes, and suggest new links to new cards whenever appropriate. Refer back to the zk docs often during the process of planning and making corrections to ensure accuracy, and if you make a mistake, figure out a way to guard against it using the zk.
4. Use `Brewfile` and `mise.toml` to install tools needed on the development workstation (typically hostnamed "gilbert", username "eblume").
5. Services are hosted either on indri directly (via ansible) or in Kubernetes (via ArgoCD). See the "Service Deployment" section below for details.
6. Try to always test changes before applying them. Use syntax checkers, do dry runs (`--check --diff`), run commands manually via `ssh indri 'some command'`, etc.
7. **Wait for user review before deploying.** After creating a PR, do not run deployment commands until the user has had a chance to review the changes. The user will indicate when they're ready to deploy.
8. After deploying changes, try to verify the result. Use `mise run indri-services-check` to do a general service health check.
## Project Structure
```
./mise-tasks/ # management and utility scripts run via `mise run`
./ansible/playbooks/ # ansible playbooks (indri.yml is primary)
./ansible/roles/ # ansible roles for indri-hosted services
Note: The user has fish abbreviations `ki` for `kubectl --context=minikube-indri` and `k9i` for `k9s --context=minikube-indri`, but these only work in interactive shells.
3. Delete the device from Tailscale admin console (user action required)
4. Deploy the k8s Ingress - it will claim the hostname
Use `ssh indri 'tailscale serve status --json'` to check current serve entries (the non-JSON output may be empty even when entries exist).
## Third-Party Projects
When a task requires cloning or using a third-party git repository (e.g., for building from source), **ask the user to mirror it on forge first**, then clone from the mirror:
This avoids external dependencies and ensures the project is available even if the upstream is unreachable.
## Task Discovery
To discover pending blumeops tasks, run:
```fish
mise run blumeops-tasks
```
This fetches tasks from the "Blumeops" project in Todoist (via 1Password for API credentials) and displays them sorted by priority: p1 (urgent), p2 (high), p4 (normal/default), p3 (backlog). The typical workflow is to pick a task from this list at the start of a session, then dive in with planning.
## Credentials
The root store for credentials is 1password, which can be accessed via `op --vault <vaultid> item get <itemid> --field fieldname --reveal`, which will prompt the user for their assent and biometrics or password. Typically, use scripts to defer this action - try not to ever grab credentials directly. For instance, the indri.yml playbook starts with `pre_tasks` to gather the relevant secrets needed to provision its services. Some services have their credentials exported to files `chmod 0600` on indri, but they still start out in 1password. In some cases you can test services with a command that grabs the credential, but try to use environment variables or other arrangements to avoid learning the credential yourself, and warn the user first.
This repo uses [Forgejo Actions](https://forgejo.org/docs/latest/user/actions/) for CI/CD. Workflows live in `.forgejo/workflows/` (not `.github/workflows/`). The runner executes jobs in host mode within the Kubernetes cluster.
Changes are classified before starting work:
## Documentation
- **C0** - quick fixes, committed directly to main
- **C1** - feature branch + PR, documentation written before code
- **C2** - multi-phase work using the Mikado method for dependency tracking
Detailed documentation lives in my personal zettelkasten, which is not included in this repository. You can view the docs with:
See the [agent change process](docs/explanation/agent-change-process.md) for
details.
```bash
mise run zk-docs
```
## License
The zettelkasten is private at time of writing. If you're interested in the documentation or have questions about this project, please reach out to blume.erich@gmail.com.
msg:"WARNING: Docker does not appear to be running. Please start Docker Desktop manually."
when:minikube_docker_status.rc != 0
- name:Check if minikube cluster exists
- name:Check minikube cluster status
ansible.builtin.command:
cmd:minikube status --format={% raw %}'{{.Host}}'{% endraw %}
cmd:minikube status
register:minikube_status
changed_when:false
failed_when:false
@ -63,11 +63,11 @@
failed_when:false# Don't fail - may need manual intervention
when:
- minikube_docker_status.rc == 0
- minikube_status.rc != 0 or 'Running' not in minikube_status.stdout
- minikube_status.rc != 0
- name:Check minikube status after start attempt
ansible.builtin.command:
cmd:minikube status --format={% raw %}'{{.Host}}'{% endraw %}
cmd:minikube status
register:minikube_final_status
changed_when:false
failed_when:false
@ -75,7 +75,38 @@
- name:Warn if minikube failed to start
ansible.builtin.debug:
msg:"WARNING: minikube may not have started properly. Run 'minikube start' manually on indri if needed. Status: {{ minikube_final_status.stdout | default('unknown') }}"
when:minikube_final_status.rc != 0 or 'Running' not in minikube_final_status.stdout
when:minikube_final_status.rc != 0
# The storage-provisioner is a bare Pod (no controller). If the node restarts
# via Docker Desktop rather than `minikube start`, kubelet brings back static
# pods (apiserver, etcd) but bare pods like storage-provisioner are lost.
# `minikube start` on a running cluster is safe and re-applies all addons.
- name:Check storage-provisioner pod is running
ansible.builtin.command:
cmd:kubectl -n kube-system get pod storage-provisioner -o jsonpath='{.status.phase}'