Add PostgreSQL functional tests with real containers in CI

.github/workflows/functional.yml

@@ -12,7 +12,7 @@ on:
 jobs:
   functional:
     runs-on: ubuntu-latest
-    timeout-minutes: 20
+    timeout-minutes: 30
 
     steps:
     - uses: actions/checkout@v4
@@ -77,17 +77,84 @@ jobs:
     - name: Run failover tests
       run: bash tests/functional/test-failover.sh
 
+    # ---- PostgreSQL functional tests ----
+    - name: Start PostgreSQL containers
+      working-directory: tests/functional
+      run: |
+        docker compose up -d pgprimary
+        echo "Waiting for pgprimary to be healthy..."
+        timeout 120 bash -c '
+          while true; do
+            STATUS=$(docker compose ps pgprimary --format json 2>/dev/null | python3 -c "
+        import json, sys
+        for line in sys.stdin:
+            svc = json.loads(line)
+            if \"healthy\" in svc.get(\"Status\",\"\").lower():
+                print(\"healthy\")
+                sys.exit(0)
+        print(\"waiting\")
+        " 2>/dev/null || echo "waiting")
+            if [ "$STATUS" = "healthy" ]; then
+              echo "pgprimary healthy"
+              exit 0
+            fi
+            sleep 2
+          done
+        ' || { echo "Timeout waiting for pgprimary"; docker compose logs pgprimary --tail=30; exit 1; }
+        docker compose up -d pgstandby1
+        echo "Waiting for pgstandby1 to be healthy..."
+        timeout 120 bash -c '
+          while true; do
+            STATUS=$(docker compose ps pgstandby1 --format json 2>/dev/null | python3 -c "
+        import json, sys
+        for line in sys.stdin:
+            svc = json.loads(line)
+            if \"healthy\" in svc.get(\"Status\",\"\").lower():
+                print(\"healthy\")
+                sys.exit(0)
+        print(\"waiting\")
+        " 2>/dev/null || echo "waiting")
+            if [ "$STATUS" = "healthy" ]; then
+              echo "pgstandby1 healthy"
+              exit 0
+            fi
+            sleep 2
+          done
+        ' || { echo "Timeout waiting for pgstandby1"; docker compose logs pgstandby1 --tail=30; exit 1; }
+
+    - name: Start PostgreSQL orchestrator
+      working-directory: tests/functional
+      run: |
+        docker compose up -d orchestrator-pg
+        echo "Waiting for PostgreSQL orchestrator to be ready..."
+        timeout 120 bash -c '
+          while true; do
+            if curl -sf http://localhost:3098/api/clusters > /dev/null 2>&1; then
+              echo "PostgreSQL orchestrator ready"
+              exit 0
+            fi
+            sleep 2
+          done
+        ' || { echo "PostgreSQL orchestrator not ready"; docker compose logs orchestrator-pg --tail=50; exit 1; }
+
+    - name: Run PostgreSQL tests
+      run: bash tests/functional/test-postgresql.sh
+
     - name: Collect orchestrator logs
       if: always()
       working-directory: tests/functional
-      run: docker compose logs orchestrator > /tmp/orchestrator-test.log 2>&1 || true
+      run: |
+        docker compose logs orchestrator > /tmp/orchestrator-test.log 2>&1 || true
+        docker compose logs orchestrator-pg > /tmp/orchestrator-pg-test.log 2>&1 || true
 
     - name: Upload orchestrator logs
       if: always()
       uses: actions/upload-artifact@v4
       with:
         name: orchestrator-test-logs
-        path: /tmp/orchestrator-test.log
+        path: |
+          /tmp/orchestrator-test.log
+          /tmp/orchestrator-pg-test.log
 
     - name: Collect all docker logs on failure
       if: failure()

tests/functional/docker-compose.yml

@@ -92,6 +92,51 @@ services:
         aliases:
           - proxysql
 
+  pgprimary:
+    image: postgres:17
+    hostname: pgprimary
+    environment:
+      POSTGRES_PASSWORD: testpass
+      POSTGRES_USER: postgres
+    volumes:
+      - ./postgres/init-primary.sh:/docker-entrypoint-initdb.d/init.sh
+    ports:
+      - "15432:5432"
+    healthcheck:
+      test: ["CMD-SHELL", "pg_isready -U postgres"]
+      interval: 5s
+      timeout: 3s
+      retries: 30
+    networks:
+      orchnet:
+        aliases:
+          - pgprimary
+
+  pgstandby1:
+    image: postgres:17
+    hostname: pgstandby1
+    environment:
+      POSTGRES_PASSWORD: testpass
+      PGUSER: postgres
+      PGPASSWORD: repl_pass
+    volumes:
+      - ./postgres/init-standby.sh:/init-standby.sh
+    entrypoint: ["/bin/bash", "/init-standby.sh"]
+    depends_on:
+      pgprimary:
+        condition: service_healthy
+    ports:
+      - "15433:5432"
+    healthcheck:
+      test: ["CMD-SHELL", "pg_isready -U postgres"]
+      interval: 5s
+      timeout: 3s
+      retries: 30
+    networks:
+      orchnet:
+        aliases:
+          - pgstandby1
+
   orchestrator:
     image: ubuntu:24.04
     hostname: orchestrator
@@ -122,6 +167,36 @@ services:
         aliases:
           - orchestrator
 
+  orchestrator-pg:
+    image: ubuntu:24.04
+    hostname: orchestrator-pg
+    volumes:
+      - ../../bin/orchestrator:/usr/local/bin/orchestrator:ro
+      - ../../resources:/orchestrator/resources:ro
+      - ./orchestrator-pg-test.conf.json:/orchestrator/orchestrator.conf.json:ro
+    command: >
+      bash -c "
+        apt-get update -qq && apt-get install -y -qq curl sqlite3 > /dev/null 2>&1 &&
+        rm -f /tmp/orchestrator-pg-test.sqlite3 &&
+        cd /orchestrator &&
+        orchestrator -config orchestrator.conf.json http
+      "
+    ports:
+      - "3098:3098"
+    depends_on:
+      pgprimary:
+        condition: service_healthy
+    healthcheck:
+      test: ["CMD", "curl", "-sf", "http://localhost:3098/api/clusters"]
+      interval: 5s
+      timeout: 3s
+      retries: 60
+      start_period: 15s
+    networks:
+      orchnet:
+        aliases:
+          - orchestrator-pg
+
 networks:
   orchnet:
     driver: bridge

tests/functional/orchestrator-pg-test.conf.json

@@ -0,0 +1,19 @@
+{
+  "Debug": true,
+  "ListenAddress": ":3098",
+  "ProviderType": "postgresql",
+  "PostgreSQLTopologyUser": "orchestrator",
+  "PostgreSQLTopologyPassword": "orch_pass",
+  "PostgreSQLSSLMode": "disable",
+  "MySQLOrchestratorHost": "",
+  "MySQLOrchestratorPort": 0,
+  "BackendDB": "sqlite",
+  "SQLite3DataFile": "/tmp/orchestrator-pg-test.sqlite3",
+  "DiscoverByShowSlaveHosts": false,
+  "InstancePollSeconds": 5,
+  "RecoveryPeriodBlockSeconds": 10,
+  "RecoverMasterClusterFilters": [".*"],
+  "RecoverIntermediateMasterClusterFilters": [".*"],
+  "AutoPseudoGTID": false,
+  "PrometheusEnabled": true
+}

tests/functional/postgres/init-primary.sh

@@ -0,0 +1,36 @@
+#!/bin/bash
+# Initialize PostgreSQL primary for functional tests.
+# This script runs inside the postgres Docker entrypoint (initdb phase).
+
+set -e
+
+# ---- WAL / replication settings ----
+cat >> "$PGDATA/postgresql.conf" <<EOF
+wal_level = replica
+max_wal_senders = 10
+wal_keep_size = 256MB
+hot_standby = on
+listen_addresses = '*'
+EOF
+
+# ---- pg_hba: allow replication and orchestrator connections ----
+cat >> "$PGDATA/pg_hba.conf" <<EOF
+# Allow replication connections from any host in the Docker network
+host    replication     repl        all     md5
+# Allow orchestrator monitoring user
+host    all             orchestrator all     md5
+EOF
+
+# ---- Create users ----
+psql -v ON_ERROR_STOP=1 --username "$POSTGRES_USER" --dbname postgres <<-EOSQL
+    -- Replication user
+    CREATE ROLE repl WITH REPLICATION LOGIN PASSWORD 'repl_pass';
+
+    -- Orchestrator monitoring user
+    CREATE ROLE orchestrator WITH LOGIN PASSWORD 'orch_pass';
+    GRANT pg_monitor TO orchestrator;
+    -- Allow orchestrator to promote standbys and reload config
+    ALTER ROLE orchestrator SUPERUSER;
+EOSQL
+
+echo "Primary initialization complete."

tests/functional/postgres/init-standby.sh

@@ -0,0 +1,37 @@
+#!/bin/bash
+# Initialize a PostgreSQL standby via pg_basebackup from the primary.
+# This script replaces the default entrypoint for standby containers.
+
+set -e
+
+PGDATA="/var/lib/postgresql/data"
+PRIMARY_HOST="pgprimary"
+PRIMARY_PORT=5432
+
+echo "Waiting for primary to accept connections..."
+until pg_isready -h "$PRIMARY_HOST" -p "$PRIMARY_PORT" -U postgres; do
+    sleep 1
+done
+
+echo "Running pg_basebackup from $PRIMARY_HOST..."
+rm -rf "$PGDATA"/*
+pg_basebackup \
+    -h "$PRIMARY_HOST" \
+    -p "$PRIMARY_PORT" \
+    -U repl \
+    -D "$PGDATA" \
+    -Fp -Xs -P -R
+
+# pg_basebackup with -R creates standby.signal and sets primary_conninfo
+# in postgresql.auto.conf. Override to ensure correct connection string.
+cat > "$PGDATA/postgresql.auto.conf" <<EOF
+primary_conninfo = 'host=$PRIMARY_HOST port=$PRIMARY_PORT user=repl password=repl_pass'
+EOF
+
+touch "$PGDATA/standby.signal"
+
+# Fix permissions (pg_basebackup may set them correctly, but be safe)
+chmod 0700 "$PGDATA"
+
+echo "Starting PostgreSQL in standby mode..."
+exec postgres -D "$PGDATA"