Prometheus

TL;DR My Mac Studio M3 Ultra runs Ollama with 70B+ models but isn’t a k3s node. I needed it to show up in Grafana next to the cluster workloads. The solution: node_exporter for system metrics, a Go reverse proxy for per-model inference metrics, a custom Python exporter for model inventory and VRAM tracking, and Grafana Alloy for shipping logs to Loki. All four services managed by Ansible, all metrics scraped by the cluster’s Prometheus. ...

TL;DR This is the moment everything was built for. Three phases of preparation — PostgreSQL provider (Day 3), storage migration (Day 4), state externalization (Day 5) — all leading to a single kubectl scale command. This post covers Phase 4: scaling the Jellyfin StatefulSet to 2 replicas, configuring anti-affinity to spread pods across nodes, running six structured failover tests, building Prometheus alerts, and one test that only partially passed. The headline result: killing a pod causes zero service downtime — users on the surviving replica experience no interruption at all, and displaced users reconnect within seconds. ...

TL;DR Grafana went completely dark for about 26 hours on my home k3s cluster. Two things broke simultaneously: Loki entered CrashLoopBackOff, and Prometheus silently stopped ingesting metrics — its pods showed as healthy and 2/2 Running the whole time. The actual cause was Longhorn’s auto-balancer migrating replicas onto a freshly-added cluster node (k3s-agent-4) that had unstable storage during its first 48 hours. The replica I/O errors propagated directly into the workloads, corrupting mid-write files: a Prometheus WAL segment and a Loki TSDB index file. Both required offline surgery via a busybox pod to delete the corrupted files before the services could recover. ...

TL;DR After running the cluster for nearly two weeks, today I took a step back to document and optimize the monitoring stack. This covers kube-prometheus-stack (Prometheus + Grafana + AlertManager), Loki for log aggregation, custom dashboards for every service, alert tuning to reduce noise, and the cluster-wide performance benchmarks I ran to establish baseline metrics. The Monitoring Architecture ┌──────────────────────────────────────────────────┐ │ Grafana │ │ ┌──────────┐ ┌──────────┐ ┌──────────┐ │ │ │ Metrics │ │ Logs │ │ Alerts │ │ │ │ Explorer │ │ Explorer │ │ Rules │ │ │ └──────┬───┘ └──────┬───┘ └──────┬───┘ │ └─────────┼──────────────┼─────────────┼───────────┘ │ │ │ ┌─────┴─────┐ ┌─────┴─────┐ │ │Prometheus │ │ Loki │ │ │ (metrics) │ │ (logs) │ │ └─────┬─────┘ └─────┬─────┘ │ │ │ ┌─────┴──────┐ ┌──────┴──────┐ ┌─────┴────┐ │AlertManager│ │ Exporters │ │Promtail │ │ → Slack │ │ node │ │(log │ └────────────┘ │ kube-state │ │ shipper) │ │ cAdvisor │ └──────────┘ │ custom │ └─────────────┘ kube-prometheus-stack The foundation is kube-prometheus-stack, deployed via Helm. This single chart installs: ...

TL;DR Today I deployed two significant additions to the cluster: an AI-powered Alert Responder that uses AWS Bedrock (Amazon Nova Micro) to analyze Prometheus alerts and post remediation suggestions to Slack, and a multi-user dev workspace with per-user environments. I also hardened the cluster by constraining all workloads to the correct architecture nodes and fixing arm64 scheduling issues. The Alert Responder Running 13+ applications on a homelab cluster means alerts fire regularly. Most are straightforward — high memory, restart loops, certificate expiry warnings — but analyzing each one, determining root cause, and knowing the right remediation command gets tedious, especially at 2 AM. ...