T05 Phase 0 Preflight Evidence — 2026-07-06

Workplan: CUST-WP-0054-T05
Parent runbook: docs/state-hub-railiance01-cutover-plan.md
no_secret_material_recorded: true

Checklist

Step	Action	Result	Evidence
0.1	Baseline API totals	pass	See § API summary below
0.2	Per-table PG counts (coulombcore)	pass	See § Table counts below
0.3	Forgejo image pull (railiance01)	pass	`main-f9f0091` + `latest` @ `sha256:0b62f2d530cf7`
0.4	Manifest dry-run (railiance01)	pass	Server + client dry-run exit 0
0.5	Storage / CNPG review	pass	`local-path` default; 75G free on `/`; CNPG healthy
0.6	Operator approval	pending	Required before Phase 2 freeze

API summary (`GET /state/summary` @ 2026-07-06T16:06:15Z)

{
  "topics": { "total": 14 },
  "workstreams": { "total": 637, "active": 11, "finished": 582 },
  "tasks": { "total": 3983, "todo": 88, "progress": 4, "done": 3696 },
  "decisions": { "total": 79, "open": 1, "resolved": 76 }
}

Full artifact: captured at preflight run from http://127.0.0.1:8000/state/summary.

Table counts — coulombcore `state-hub-db` (production writer)

Queried via kubectl exec on databases/state-hub-db-1 as postgres:

Table	Count	CUST-WP-0011-T07 reference
`workplans`	637	was 633 at 2026-07-03 cutover
`tasks`	3983	was 3964
`progress_events`	8717	was 8192
`topics`	14	was 14
`token_events`	1936	was 1933
`managed_repos`	74	—
`decisions`	79	—

Cutover gate: post-restore counts on railiance01 must match this table exactly.

Production deployment (coulombcore, source of dump)

Item	Value
API image	`gitea.coulomb.social/coulomb/state-hub:f2e042a`
API pod	`state-hub/state-hub-7d565cddf4-jnpfw` Running
CNPG cluster	`databases/state-hub-db` — healthy, 11d
ClusterIP	`10.43.170.94:8000`
Workstation access	ops-bridge `state-hub-primary` → above Svc

Target image (railiance01 deploy)

Tag	Digest	Notes
`forgejo.coulomb.social/coulomb/state-hub:main-f9f0091`	`sha256:0b62f2d530cf7`	CI build from tier-3 promotion
`forgejo.coulomb.social/coulomb/state-hub:latest`	same	Alias
`:f9f0091`	not found	Use `main-f9f0091` or `latest` in Helm values

railiance01 platform readiness

Check	Status
CNPG operator (`cnpg-system`)	Running
Existing CNPG clusters	`forgejo-db`, `net-kingdom-pg` healthy
`state-hub` namespace	Does not exist (expected — Phase 1 creates)
`state-hub-db` on railiance01	Not deployed (expected)
StorageClass	`local-path` (default)
Node disk	75G available on `/` (96G volume, 23% used)
Sweep hostname	`239.62.205.92.host.secureserver.net`
Partial clone tree	`/home/tegwick/activity-core`, `llm-connect`, …

Dry-run output summary

cd ~/state-hub
KUBECONFIG=~/.kube/config-hosteurope make railiance-state-hub-server-dry-run  # exit 0
KUBECONFIG=~/.kube/config-hosteurope make railiance-state-hub-client-dry-run   # exit 0

Server dry-run created (dry-run): state-hub-db-credentials, state-hub-db Cluster, DB NetworkPolicies, state-hub Namespace, state-hub-env, ConfigMap, Service, Deployment.

Phase 1 readiness

Phase 0 is complete except operator approval (0.6). Phase 1 may proceed without approval (empty stack deploy). Phase 2 freeze/restore requires explicit approval.

Recommended Helm image pin for Phase 1

image:
  repository: forgejo.coulomb.social/coulomb/state-hub
  tag: "main-f9f0091"

Update state-hub/deploy/railiance/apps/helm/state-hub-values.yaml and promote to railiance-apps before or during Phase 1.

Open items carried to Phase 1

Create SOPS-backed state-hub-db-credentials on railiance01 (platform)
Create state-hub-env Secret with DATABASE_URL pointing at railiance01 CNPG rw svc
Confirm image pull secret if Forgejo registry requires auth from cluster (currently pulls anonymously)

4.1 KiB Raw Blame History