- NodeDown now suppresses workload and service alerts (PodCrashLooping, DeploymentReplicasMismatch, StatefulSetReplicasMismatch, etc.) - NFSServerUnresponsive suppresses pod-level alerts - Increased for durations on transient alerts (e.g. 15m→30m for replica mismatches) - NodeDown for: 1m→3m to avoid flapping - Removed all 3 Loki log-based alerts (duplicated Prometheus alerts) - Downgraded HeadscaleDown critical→warning, mail server page→warning |
||
|---|---|---|
| .. | ||
| modules | ||
| .gitkeep | ||
| .terraform.lock.hcl | ||
| backend.tf | ||
| main.tf | ||
| providers.tf | ||
| redis-25.3.2.tgz | ||
| secrets | ||
| terragrunt.hcl | ||
| tiers.tf | ||