Merge branch 'chore/bump-flask-resource-limits' into 'main'

chore: bump flask backend resource limits (CPU 0.5, mem 1Gi)

Closes #168

See merge request postgres-ai/postgresai!238

Author: Sarumyan

.gitlab-ci.yml

@@ -212,7 +212,9 @@ vm-auth:config:tests:
     - helm repo add grafana https://grafana.github.io/helm-charts
     - helm dependency build postgres_ai_helm/
   script:
-    - python -m pytest tests/compliance_vectors/test_vm_auth.py -v
+    - helm lint postgres_ai_helm/
+    - helm template test postgres_ai_helm/ --set secrets.createFromValues=true >/tmp/postgres-ai-helm-rendered.yaml
+    - python -m pytest tests/compliance_vectors/test_vm_auth.py tests/compliance_vectors/test_flask_resources.py -v
   rules:
     - if: '$CI_PIPELINE_SOURCE == "merge_request_event"'
     - if: '$CI_COMMIT_BRANCH == "main"'

docker-compose.yml

@@ -246,8 +246,8 @@ services:
   monitoring_flask_backend:
     image: ${PGAI_REGISTRY:-postgresai}/monitoring-flask-backend:${PGAI_TAG:?PGAI_TAG is required}
     container_name: flask-pgss-api
-    cpus: 0.1
-    mem_limit: 192m
+    cpus: 0.5
+    mem_limit: 1073741824
     environment:
       - FLASK_ENV=production
       - PROMETHEUS_URL=http://sink-prometheus:9090

postgres_ai_helm/values.yaml

@@ -76,7 +76,13 @@ flask:
   service:
     type: ClusterIP
     port: 8000
-  resources: {}
+  resources:
+    requests:
+      cpu: 500m
+      memory: 256Mi
+    limits:
+      cpu: 500m
+      memory: 1Gi
 
 reporter:
   enabled: true

tests/compliance_vectors/test_flask_resources.py

@@ -0,0 +1,73 @@
+"""Resource contract tests for the monitoring Flask backend."""
+
+import os
+import subprocess
+
+import pytest
+import yaml
+
+
+PROJECT_ROOT = os.path.abspath(os.path.join(os.path.dirname(__file__), '..', '..'))
+EXPECTED_MEMORY_BYTES = 1073741824
+
+
+def _load_yaml(*parts):
+    with open(os.path.join(PROJECT_ROOT, *parts)) as f:
+        return yaml.safe_load(f)
+
+
+def test_docker_compose_flask_resources_match_expected_limits():
+    compose = _load_yaml('docker-compose.yml')
+    flask = compose['services']['monitoring_flask_backend']
+
+    assert flask['cpus'] == 0.5
+    assert flask['mem_limit'] == EXPECTED_MEMORY_BYTES
+
+
+def test_helm_flask_resources_match_expected_requests_and_limits():
+    values = _load_yaml('postgres_ai_helm', 'values.yaml')
+    resources = values['flask']['resources']
+
+    assert resources['requests'] == {
+        'cpu': '500m',
+        'memory': '256Mi',
+    }
+    assert resources['limits'] == {
+        'cpu': '500m',
+        'memory': '1Gi',
+    }
+
+
+@pytest.fixture
+def rendered_helm_docs():
+    try:
+        subprocess.run(['helm', 'version', '--short'], capture_output=True, check=True)
+    except (FileNotFoundError, subprocess.CalledProcessError):
+        pytest.skip('helm not available')
+
+    chart_path = os.path.join(PROJECT_ROOT, 'postgres_ai_helm')
+    subprocess.run(['helm', 'dependency', 'build', chart_path], capture_output=True, check=False)
+    result = subprocess.run(
+        ['helm', 'template', 'test', chart_path, '--set', 'secrets.createFromValues=true'],
+        capture_output=True,
+        text=True,
+    )
+    if result.returncode != 0:
+        pytest.fail(f'helm template failed: {result.stderr}')
+
+    return [doc for doc in yaml.safe_load_all(result.stdout) if isinstance(doc, dict)]
+
+
+def test_helm_template_renders_flask_resources(rendered_helm_docs):
+    flask_deployments = [
+        doc for doc in rendered_helm_docs
+        if doc.get('kind') == 'Deployment'
+        and doc.get('metadata', {}).get('name', '').endswith('-flask')
+    ]
+    assert len(flask_deployments) == 1
+
+    container = flask_deployments[0]['spec']['template']['spec']['containers'][0]
+    assert container['resources'] == {
+        'requests': {'cpu': '500m', 'memory': '256Mi'},
+        'limits': {'cpu': '500m', 'memory': '1Gi'},
+    }