K3S-HOME/observability
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
176 Commits 1 Branch 0 Tags
4bdcaf8fcd4967e25bf1beb4d74c43277ac4927b
Commit Graph

10 Commits

Author SHA1 Message Date
Mayne0213
203a8debac REFACTOR(repo): remove control-plane scheduling 
- Remove nodeSelector for control-plane node
- Remove tolerations for control-plane taint
- Allow pods to schedule on any available node
 2026-01-10 18:35:15 +09:00
Mayne0213
a3003d597f PERF(observability): adjust resources based on VPA 
- Update blackbox-exporter cpu 15m→23m, memory 64Mi→100Mi
- Update grafana cpu 11m→23m, memory 425Mi→175Mi
- Update loki cpu 23m→63m, memory 462Mi→363Mi
- Update tempo cpu 50m→15m, memory 128Mi→100Mi
- Update thanos memory 128Mi→283Mi
- Update node-exporter memory 64Mi→100Mi
- Update kube-state-metrics memory 100Mi→105Mi
- Update opentelemetry-operator cpu 10m→11m, memory 256Mi→75Mi
- Update vpa memory 128Mi→100Mi
 2026-01-10 14:33:40 +09:00
Mayne0213
5089e8607d CHORE(resources): set memory limits equal to memory requests 
Align memory limits with memory requests for guaranteed QoS class.
- prometheus, thanos (query, storegateway, compactor)
- alertmanager, tempo, goldilocks (dashboard, controller)
- node-exporter, opentelemetry-collector, vpa, kube-state-metrics
 2026-01-09 21:42:35 +09:00
Mayne0213
735166fc9c REFACTOR(repo): standardize taint to control-plane 
- Change node-role.kubernetes.io/master to control-plane
- Update vpa, goldilocks, kube-state-metrics tolerations
- Remove deprecated master taint from promtail
 2026-01-09 21:41:52 +09:00
Mayne0213
4511fd5b2e FIX(repo): correct nodeSelector label value 
- Change master label value from "" to "true"
- Fix pod scheduling failure due to label mismatch
 2026-01-09 21:41:52 +09:00
Mayne0213
1c6a9dc491 PERF(repo): move system pods to master node 
- Add nodeSelector for master node placement
- Add tolerations for NoExecute taint
- kube-state-metrics: schedule on master
- goldilocks-controller: schedule on master, reduce to 1 replica
- vpa-recommender: schedule on master, remove anti-affinity
- Free worker node resources for applications
 2026-01-09 21:41:52 +09:00
Mayne0213
4515ea0b33 FEAT(observability): enable HA with replica 2 and soft anti-affinity 
- Add replicaCount: 2 to goldilocks, vpa, alertmanager
- Add replicas: 2 to loki singleBinary
- Add soft pod anti-affinity for node distribution
- Keep kube-state-metrics at replica 1 to prevent duplicate metrics

FIX(loki): revert to replica 1 for Single Binary mode

- Single Binary mode cannot run more than 1 replica without object storage
- Remove affinity configuration for single replica
- Keep filesystem storage backend
 2026-01-09 21:41:51 +09:00
Mayne0213
4286296591 PERF(resources): remove CPU limits - keep memory limits only 
- CPU throttling prevents app startup, not crashes
- Memory OOM is the real cascading failure cause
- CPU request ensures fair scheduling
 2026-01-07 23:48:35 +09:00
Mayne0213
ac2abde8b5 FIX(prometheus): servicemonitor namespace 
- from monitoring to prometheus
 2026-01-04 23:38:05 +09:00
Mayne0213
5c4676ca9a REFACTOR(repo): restructure monitoring folder 
- and add namespace resou...
- Remove argocd/, helm-values/, ingress/ subdirectories
- Move files to parent directory (argocd.yaml, helm-values.yaml,
  ingress.yaml)
- Update helm valueFiles paths in ArgoCD Applications
- Add namespace.yaml to all applications with Goldilocks labels
- Update destination namespaces to match folder names
- Update kustomization.yaml files to reference new structure
 2026-01-04 23:38:05 +09:00