73512e92a6
- Updated deployment with correct Pi 3 tolerations - Updated ingress for cloudflare-tunnel - Added crontab example for systemd alternative - Updated go.sum 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
35 lines
884 B
Markdown
35 lines
884 B
Markdown
# Cluster Issue Diagnosis Plan
|
|
|
|
## Issues to Investigate
|
|
|
|
1. **Critical Alerts** - KubeSchedulerDown, KubeControllerManagerDown
|
|
- Likely false positives (k0s bundles these in k0s-controller)
|
|
- Check if cluster is actually functional
|
|
|
|
2. **CrashLooping Pod** - Find and diagnose
|
|
- Get pod status across all namespaces
|
|
- Check logs and events
|
|
|
|
3. **Stuck Deployment** - Find and diagnose
|
|
- List deployments not at desired replica count
|
|
- Check events
|
|
|
|
4. **Degraded kube-prometheus-stack**
|
|
- Check prometheus/alertmanager pods
|
|
|
|
## Commands to Run
|
|
|
|
```bash
|
|
# Find crash looping pods
|
|
kubectl get pods -A | grep -E 'CrashLoop|Error|ImagePull'
|
|
|
|
# Find stuck deployments
|
|
kubectl get deploy -A -o wide | grep -v '1/1\|2/2\|3/3\|4/4'
|
|
|
|
# Check prometheus stack
|
|
kubectl get pods -n monitoring
|
|
|
|
# Check scheduler/controller (k0s specific)
|
|
kubectl get pods -n kube-system
|
|
```
|