Introduction When migrating monitoring systems, losing historical alert data and baselines creates a blind spot for anomaly detection. Teams cannot compare current metrics against historical trends.

Symptoms - No historical data in new monitoring system - Alert rules not migrated - Dashboards missing panels or showing empty data - Anomaly detection not working without baseline - Cannot compare metrics before and after migration

Common Causes - Historical data not exported from old system - Alert rules not recreated in new system - Metric naming convention changed - Dashboard definitions not migrated - Data retention period reset

Step-by-Step Fix 1. **Export historical data from old system': ```bash # Prometheus remote write to new system # Grafana dashboard export curl -s http://grafana/api/dashboards/uid/<uid> > dashboard.json ```

  1. 1.**Recreate alert rules in new system':
  2. 2.Document and recreate all alert rules with equivalent thresholds.
  3. 3.**Backfill historical data if possible':
  4. 4.Use remote read/write to transfer historical metrics.

Prevention - Plan monitoring migration as a phased process - Run old and new monitoring systems in parallel - Export all alert rules and dashboards before decommissioning - Maintain metric naming compatibility - Document all monitoring baselines before migration