pgwatch generates excessive INFO logs (~80% of namespace log volume)

Problem

The pgwatch container in postgres-ai-mon-production is generating ~80% of all logs in the namespace due to verbose INFO-level logging.

Evidence 

$ gcloud logging read 'resource.type="k8s_container" AND resource.labels.namespace_name="postgres-ai-mon-production"' \
  --limit=10000 --format='value(resource.labels.container_name)' | sort | uniq -c | sort -nr

8138 pgwatch       # 81% of all logs
1509 grafana
 306 postgres-exporter
  47 victoriametrics

Root Cause

pgwatch logs every metric fetch at INFO level:

[INFO] [source:prod_replica_postgres_ai] [metric:replication] [rows:0] measurements fetched
[INFO] [source:prod_replica_postgres_ai] [metric:pg_stat_replication] [rows:0] measurements fetched
[INFO] [source:prod_replica_postgres_ai] [metric:pg_blocked] [rows:0] measurements fetched
...

With 4 database sources × ~25 metrics × 30-second intervals = ~12,000 log lines per hour of noise.

Solution

Add --log-level=warn to pgwatch services to suppress INFO-level logs while keeping warnings and errors.

Impact

  • Reduced log storage costs in GCP Logging
  • Easier to find actual issues in logs
  • No impact on metrics collection or monitoring functionality