feat(2): Q4.6 discourse — recipe_meta + postgres P4 overlays + health (WIP, §4.3 create-topic next)

discourse (forum: postgres+redis+sidekiq). HEALTH_PATH=/srv/status (slow Rails boot, DEPLOY_TIMEOUT=1800).
P4 via postgres ci_marker (db service, pg_dump backupbot — matrix-synapse pattern). Health functional
test. §4.3 create-a-topic + PARITY.md to follow after smoke discovers the admin/API bootstrap path.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

tests/discourse/functional/test_health_check.py

@@ -0,0 +1,20 @@
+"""discourse — health/readiness functional test (Phase 2).
+
+SOURCE: no recipe-maintainer corpus exists for discourse (P2 vacuous — see PARITY.md). This is a
+Phase-2 health check aligned with the parity-port convention: /srv/status returns 200 only once the
+Rails app is actually serving (the recipe's HEALTH_PATH and the canonical "is discourse up" signal).
+"""
+
+from __future__ import annotations
+
+import os
+import sys
+
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", "..", "..", "runner"))
+from harness import http as harness_http  # noqa: E402
+
+
+def test_discourse_srv_status_ok(live_app):
+    url = f"https://{live_app}/srv/status"
+    status, _ = harness_http.retry_http_get(url, expect_status=200, max_wait=120, interval=5)
+    assert status == 200, f"GET {url} HTTP {status} (expected 200 — discourse not serving)"

tests/discourse/ops.py

@@ -0,0 +1,47 @@
+"""discourse — pre-op seed hooks (Phase 1e HC3 / Phase 2 P4 backup data-integrity).
+
+Discourse persists in postgres (`db` service, user/db=discourse). The recipe's backupbot pre-hook
+pg_dumps that DB, and restore reloads the dump — so real backup data-integrity = seed a `ci_marker`
+row into the discourse DB, back up, mutate, restore, and prove the row survived (the same DB the
+recipe actually backs up). Mirrors the matrix-synapse postgres pattern.
+"""
+
+import os
+import sys
+
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", "..", "runner"))
+from harness import lifecycle  # noqa: E402
+
+
+def _psql(domain, sql):
+    cmd = (
+        'PGPASSWORD=$(cat /run/secrets/db_password) '
+        f'psql -U discourse -d discourse -tAc "{sql}"'
+    )
+    return lifecycle.exec_in_app(domain, ["sh", "-c", cmd], service="db").strip()
+
+
+def _seed(domain, value):
+    _psql(
+        domain,
+        "CREATE TABLE IF NOT EXISTS ci_marker(v text); DELETE FROM ci_marker; "
+        f"INSERT INTO ci_marker VALUES('{value}');",
+    )
+    got = _psql(domain, "SELECT v FROM ci_marker;")
+    assert got == value, f"seed did not commit (read back {got!r}, expected {value!r})"
+
+
+def pre_upgrade(domain, meta):
+    _seed(domain, "upgrade-survives")
+
+
+def pre_backup(domain, meta):
+    _seed(domain, "original")
+
+
+def pre_restore(domain, meta):
+    # diverge from the backup so a successful restore is observable
+    _psql(domain, "DROP TABLE IF EXISTS ci_marker;")
+    assert _psql(domain, "SELECT to_regclass('public.ci_marker');") in ("", "NULL"), (
+        "drop did not take"
+    )

tests/discourse/recipe_meta.py

@@ -0,0 +1,11 @@
+# Per-recipe harness config for discourse (Phase 2 Q4.6 — forum; postgres + redis + sidekiq).
+#
+# Discourse (bitnami/discourse) is a slow-booting Rails app: the recipe healthcheck polls
+# /srv/status with a 5-minute start_period, and a cold first boot (DB migrate + asset precompile)
+# regularly takes 8-15 min, so the deploy/HTTP timeouts are generous. /srv/status returns 200 only
+# once the app is actually serving (the canonical "is discourse up" signal — NOT "/", which may
+# redirect to setup).
+HEALTH_PATH = "/srv/status"
+HEALTH_OK = (200,)
+DEPLOY_TIMEOUT = 1800
+HTTP_TIMEOUT = 1200

tests/discourse/test_backup.py

@@ -0,0 +1,26 @@
+"""discourse — BACKUP overlay (Phase 1e HC3 / Phase 2 P4): assertion-only + additive.
+
+ops.pre_backup seeded ci_marker='original' into the discourse postgres DB (which the recipe's
+backupbot pre-hook pg_dumps). This overlay ADDS: the seeded row is intact at backup time. The
+backup→restore divergence (dropping the table) is in ops.pre_restore.
+"""
+
+import os
+import sys
+
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", "..", "runner"))
+from harness import lifecycle  # noqa: E402
+
+
+def _psql(domain, sql):
+    cmd = (
+        'PGPASSWORD=$(cat /run/secrets/db_password) '
+        f'psql -U discourse -d discourse -tAc "{sql}"'
+    )
+    return lifecycle.exec_in_app(domain, ["sh", "-c", cmd], service="db").strip()
+
+
+def test_backup_captures_state(live_app):
+    assert _psql(live_app, "SELECT v FROM ci_marker;") == "original", (
+        "the seeded discourse postgres state was not present at backup time"
+    )

tests/discourse/test_restore.py

@@ -0,0 +1,26 @@
+"""discourse — RESTORE overlay (Phase 1e HC3 / Phase 2 P4): data-integrity, assertion-only.
+
+ops.pre_restore dropped the ci_marker table (diverge); the recipe's restore reloads the pg_dump.
+This overlay ADDS: the restored DB carries the pre-mutation 'original' marker — proving the seeded
+data survived backup→restore, not just that the service came back up.
+"""
+
+import os
+import sys
+
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", "..", "runner"))
+from harness import lifecycle  # noqa: E402
+
+
+def _psql(domain, sql):
+    cmd = (
+        'PGPASSWORD=$(cat /run/secrets/db_password) '
+        f'psql -U discourse -d discourse -tAc "{sql}"'
+    )
+    return lifecycle.exec_in_app(domain, ["sh", "-c", cmd], service="db").strip()
+
+
+def test_restore_returns_state(live_app):
+    assert _psql(live_app, "SELECT v FROM ci_marker;") == "original", (
+        "restore did not return the pre-mutation discourse postgres state (data-integrity failure)"
+    )