QUALITY_AND_MONITORING.md

Quality and monitoring (index)

Operational checks split by concern so paths match how tools run:

• Automated tests and dbt Data Quality: TESTING_AND_DATA_QUALITY.md — pytest layouts, dbt/tests, selector dqc_all_tests, dwh_dq, scripts/run_dqc.sh.
• Grafana/Prometheus/meta/logging: OBSERVABILITY_AND_LOGGING.md — canonical dashboards under infra/monitoring/grafana/, meta.* views, metrics.

Quick commands: make smoke, make p0-verify, make dqc-help. После подъёма стека с хоста: scripts/ingress_smoke.sh (портал, Superset/nl2sql health, Prometheus за ingress).

Contracts (compact)

• Ingress: /, /dbt/ (dbt Docs), /airflow/, /mlflow/, /grafana/.
• meta.pipeline_runs: status values running, success, failed; finished_at when a run completes (see services/common/run_metadata.py).
• Grafana: Prometheus panels from Pushgateway (dag_*, dataops_dbt_*, task_duration_seconds_*); SQL panels from meta.v_pipeline_runs_recent and meta.v_dq_recent via datasource PostgreSQL DWH (uid: PGDWH).

Debugging flow

1. URLs: make smoke.
2. dbt / DQ failures: ./scripts/run_dqc.sh or manual dbt test --selector dqc_all_tests; inspect dwh_dq for stored failures.
3. meta.* views and Meta telemetry (Postgres) dashboard in Grafana (see OBSERVABILITY doc).