goldmane-trail: polish follow-ups #57/#59/#61/#62/#63 + digest→#alerts
All checks were successful
ci/woodpecker/push/default Pipeline was successful

Completes the Goldmane who-talks-to-whom trail (ADR-0014), implemented by a
subagent workflow (distinct stacks in parallel, docs last):

- #57 Whisker gated ingress: ingress_factory (whisker.viktorbarzin.me,
  auth=required, Authentik-gated) + a NetworkPolicy allowing traefik->whisker:8081
  (the operator's whisker NP default-denies ingress). calico stack.
- #61 pipeline health: AggregatorDown + DigestFailing Prometheus alerts
  (prometheus_chart_values.tpl) + cluster-health check #48.
- #59 service-identity labels on the multi-Service namespaces (monitoring's 5
  TF-managed deployments + dbaas), with the KYVERNO_LIFECYCLE_V1 marker so they
  update in-place.
- #62/#63 docs: docs/runbooks/goldmane-flow-trail.md (new), service-catalog,
  security.md + monitoring.md east-west sections, ADR-0014 as-built, CONTEXT.md.
  #62 = the SQL to derive the Wave-1 per-namespace egress allowlist from the
  edge table (feeds code-8ywc; enforce-flips out of scope).

Also fixes the digest's Slack target: #security override 404s channel_not_found
because the shared alertmanager_slack_api_url webhook's app isn't a member of
#security (this likely also breaks alertmanager's slack-security receiver — flagged
in the runbook). Routed to #alerts (the webhook's working channel) until the app
is invited; verified a real digest run posts cleanly (360 edges).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
Viktor Barzin 2026-06-25 17:49:25 +00:00
parent 306cdd4cb3
commit 6c5288998f
17 changed files with 626 additions and 11 deletions

View file

@ -13,6 +13,8 @@
| authentik | Identity provider (SSO) | authentik |
| cloudflared | Cloudflare tunnel | cloudflared |
| authelia | Auth middleware (may be merged into ebooks or removed) | platform |
| goldmane | Calico 3.30 OSS flow aggregator (`goldmane.calico-system.svc:7443`, gRPC/mTLS). Stamps identity (ns/pod/workload/labels + allow-deny) on every flow from Felix into a ~60-min in-memory ring buffer — no etcd/API writes. East-west "who-talks-to-whom" source (ADR-0014). Enabled via operator CR (`kubectl_manifest.goldmane`). | calico |
| whisker | Calico 3.30 OSS live flow-observability UI (`whisker.calico-system.svc:8081`) at `whisker.viktorbarzin.me` (Authentik-gated, `auth=required` — no own login; additive NP ORs Traefik past the operator default-deny). ~60-min live view of Goldmane flows, NOT history. Enabled via operator CR (`kubectl_manifest.whisker`). | calico |
| monitoring | Prometheus/Grafana/Loki stack | monitoring |
## Storage & Security (Tier: cluster)
@ -37,6 +39,7 @@
## Active Use
| Service | Description | Stack |
|---------|-------------|-------|
| goldmane-edge-aggregator | Durable who-talks-to-whom audit trail (ADR-0014 / #58). Go service: `aggregate` Deployment streams Goldmane's gRPC `Flows.Stream` (mTLS) and upserts the low-cardinality namespace-pair edge set (`edge(src_ns,dst_ns,action,first_seen,last_seen,flow_count)`) into CNPG DB `goldmane_edges`; `goldmane-edges-digest` CronJob posts first-seen edges daily to `#security`. mTLS client cert REUSES the operator's `whisker-backend-key-pair` (re-apply if rotated). Tier-4-aux. Image `ghcr.io/viktorbarzin/goldmane-edge-aggregator` (private). Runbook: [goldmane-flow-trail.md](../../docs/runbooks/goldmane-flow-trail.md). | goldmane-edge-aggregator |
| mailserver | Email (docker-mailserver) | mailserver |
| shadowsocks | Proxy | shadowsocks |
| webhook_handler | Webhook processing | webhook_handler |
@ -161,3 +164,4 @@ procedures) are documented in `infra/docs/runbooks/`:
| pfSense + Unbound DNS | [pfsense-unbound.md](../../docs/runbooks/pfsense-unbound.md) |
| Mailserver PROXY-protocol / HAProxy | [mailserver-pfsense-haproxy.md](../../docs/runbooks/mailserver-pfsense-haproxy.md) |
| Technitium apply flow | [technitium-apply.md](../../docs/runbooks/technitium-apply.md) |
| Goldmane flow trail (east-west who-talks-to-whom) | [goldmane-flow-trail.md](../../docs/runbooks/goldmane-flow-trail.md) |