rapp-canned-prompts/manifests/database-secrets.yaml
tegwick 4d5af7698d First deployment: verified on railiance01
RCP-WP-0002 T02-T04 done, readiness_state verified with evidence attached
rather than ahead of it.

rapp-postgres provisioned canned_prompts on platform-pg-2 and sent a
database-owner receipt with 12 checks proven live. creds/canned-prompts-publish
was deliberately not issued, so the service runs read-only and POST /packages
returns 503 explaining why — the intended posture, not a gap.

Four defects surfaced that only a real rollout could expose, two of them
silent:

- SET ROLE opened an implicit transaction that Alembic nested inside rather
  than owning, so every revision logged as applied and was rolled back.
  Alembic reported success against an empty database.
- The egress NetworkPolicy selected app.kubernetes.io/name, which the
  migration Job does not carry. The Job matched only the default-deny and
  succeeded exactly once, because it ran before the policies existed; the next
  migration would have failed on DNS. Now selects part-of, with ingress split
  into its own policy so the Job is never reachable.
- env.py read database_url rather than resolved_database_url, so the migration
  could never run where the credential is a mounted file.
- live-image-digest-match extracted the pin with a line-offset grep, which
  returned empty once comments were added above `version:`. The check degraded
  to reporting "not pinned yet" while a digest was pinned — it could not have
  passed for any pin. Now parsed as YAML. A verification step that cannot fail
  is worth less than none, because it is trusted.

Evidence: docs/evidence/RCP-WP-0002-T04-first-deployment-2026-09-08.md

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bjefh8NUiEiahN4JLwoSKM

Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 388925@bnt-lap001
Assistant-Session: 3507023f-e0fd-4a1e-9d90-a0d4217d1502
2026-09-08 08:56:02 +02:00

87 lines
2.9 KiB
YAML

# Two credentials, deliberately. The runtime role can read and write rows; the
# migration role owns the schema. A service that can ALTER its own tables at
# runtime turns any code defect into a schema defect.
apiVersion: external-secrets.io/v1
kind: ClusterSecretStore
metadata:
name: openbao-canned-prompts-database
labels:
app.kubernetes.io/name: canned-prompts
railiance-platform/component: external-secrets
spec:
conditions:
- namespaces:
- canned-prompts
provider:
vault:
server: http://openbao.openbao.svc:8200
path: database
version: v1
auth:
tokenSecretRef:
name: openbao-canned-prompts-eso-token
namespace: external-secrets
key: token
---
apiVersion: external-secrets.io/v1
kind: ExternalSecret
metadata:
name: canned-prompts-postgres-runtime
namespace: canned-prompts
spec:
refreshInterval: 5m
secretStoreRef:
kind: ClusterSecretStore
name: openbao-canned-prompts-database
target:
name: canned-prompts-postgres-runtime
creationPolicy: Owner
deletionPolicy: Retain
template:
engineVersion: v2
data:
url: >-
postgresql+psycopg://{{ .username | urlquery }}:{{ .password | urlquery }}@platform-pg-2-rw.databases.svc.cluster.local:5432/canned_prompts?sslmode=require
dataFrom:
- extract:
key: creds/canned-prompts-runtime
---
apiVersion: external-secrets.io/v1
kind: ExternalSecret
metadata:
name: canned-prompts-postgres-migration
namespace: canned-prompts
spec:
refreshInterval: 5m
secretStoreRef:
kind: ClusterSecretStore
name: openbao-canned-prompts-database
target:
name: canned-prompts-postgres-migration
creationPolicy: Owner
deletionPolicy: Retain
template:
engineVersion: v2
data:
url: >-
postgresql+psycopg://{{ .username | urlquery }}:{{ .password | urlquery }}@platform-pg-2-rw.databases.svc.cluster.local:5432/canned_prompts?sslmode=require
dataFrom:
- extract:
key: creds/canned-prompts-migration
---
# The publish token is deliberately NOT declared as an ExternalSecret.
#
# rapp-postgres confirmed on 2026-09-08 that `creds/canned-prompts-publish`
# does not exist and is not being created, at our request. An ExternalSecret
# pointing at it would sit permanently unresolved — indistinguishable from a
# broken deployment, and an invitation for the next operator to "fix" it by
# minting a credential nobody decided to issue.
#
# Without the secret the service runs READ-ONLY, which is the intended posture
# until per-publisher identity exists upstream (RCP-WP-0002-T05). The runtime
# Deployment mounts the secret `optional: true`, so it starts cleanly without it
# and picks the token up if one is ever added.
#
# To enable publishing later: have rapp-postgres issue
# `creds/canned-prompts-publish`, add the ExternalSecret here, and revisit the
# NetworkPolicy ingress rule, which currently admits any namespace.