47640511c4
Add Loki metrics scraping and Grafana dashboard
...
- Add Loki scrape job to Prometheus config
- Add Loki dashboard with storage/ingestion metrics:
- Chunks in memory
- Active chunks/streams
- Heap usage
- Ingestion rate (bytes/sec)
- Log lines rate
- Memory usage over time
- Storage growth (24h rolling)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-15 12:14:32 -08:00
8291a8cb22
Remove ordering comments from playbook (handled by meta dependencies)
2026-01-15 12:02:53 -08:00
7fd601d864
Add role dependencies via meta/main.yml
...
- alloy depends on prometheus and loki
- grafana depends on prometheus and loki
- loki has no dependencies
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-15 12:01:18 -08:00
09c432e0c1
Add Grafana Alloy and Loki for unified observability
...
- Add ansible/roles/alloy/ - replaces node_exporter for metrics collection
- Uses prometheus.exporter.unix with textfile collector
- Pushes metrics to Prometheus via remote_write
- Collects logs from brew services and mcquack LaunchAgents
- Forwards logs to Loki
- Add ansible/roles/loki/ - log storage and query engine
- Single-node filesystem-based deployment
- TSDB storage with 31-day retention
- Integrated with Grafana as datasource
- Update Prometheus to enable remote_write receiver
- Remove node-exporter-indri scrape job (Alloy pushes instead)
- Keep sifaka scraping via traditional node_exporter
- Update Grafana datasources to include Loki
- Update indri-services-check to verify Loki and Alloy
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-01-15 11:51:50 -08:00
