⚡ SignalPilot RCA Report

Namespace: signalpilot-test | Generated: 2026-06-03 18:43 UTC | Analysis ID: a9739fde-c1b | Duration: 0.0s
1
Critical
2
High
0
Medium
3
Total Findings
4
Sources

Analysis of namespace 'signalpilot-test' completed at 2026-06-03 18:43 UTC.

SignalPilot identified 3 finding(s): 1 critical, 2 high. Sources analyzed: kube_api, events, metrics_server, logs. Analysis completed in 0.0s.

Top finding [CRITICAL] 'ImagePullBackOff — container image cannot be pulled' (confidence 98%): Kubernetes cannot pull the container image. Possible causes: wrong tag, registry auth failure, network issue, or image doesn't exist. Recommended first action: Verify the image tag exists in the registry. Run: `kubectl describe pod -n signalpilot-test -l app=sp-test-imagepull`

🔍 Findings (3)
critical ImagePullBackOff — container image cannot be pulled conf 98% | blast 8% | Pod/sp-test-imagepull-84d9d58795-zpfvt /app
Kubernetes cannot pull the container image. Possible causes: wrong tag, registry auth failure, network issue, or image doesn't exist.

🔧 Recommended Fixes

1. Verify the image tag exists in the registry
kubectl describe pod -n signalpilot-test -l app=sp-test-imagepull
2. Roll back to previous image
kubectl rollout undo deployment/sp-test-imagepull

📊 Evidence (1 source(s))

[SignalSource.KUBE_API] SignalKind.IMAGE_PULL_ERROR | sp-test-imagepull-84d9d58795-zpfvt/app | Container app: ImagePullBackOff
high Readiness/liveness probe failing — service endpoints not ready conf 80% | blast 8% | Pod/sp-test-imagepull-84d9d58795-zpfvt
Pod sp-test-imagepull-84d9d58795-zpfvt probe is failing, causing it to be removed from Service endpoints. Check probe path, port, timing parameters, and whether the application is actually healthy.

🔧 Recommended Fixes

1. Inspect probe configuration
kubectl describe pod sp-test-imagepull-84d9d58795-zpfvt -n signalpilot-test
2. Increase probe initialDelaySeconds if app needs more startup time
livenessProbe: initialDelaySeconds: 60 # increase from current value periodSeconds: 10 failureThreshold: 3

📊 Evidence (1 source(s))

[SignalSource.KUBE_API] SignalKind.PROBE_FAILURE | sp-test-imagepull-84d9d58795-zpfvt | ContainersReady=False reason=ContainersNotReady
high Pod unschedulable — insufficient cluster capacity conf 70% | blast 8% | Pod/sp-test-imagepull-84d9d58795-zpfvt
Pod sp-test-imagepull-84d9d58795-zpfvt is stuck in Pending state. The cluster may not have sufficient CPU/memory to schedule this pod.

🔧 Recommended Fixes

1. Check node capacity and pod resource requests
kubectl describe nodes | grep -A5 'Allocated resources'
2. Reduce CPU/memory requests if over-requested
kubectl set resources deployment/sp-test-imagepull --requests=cpu=100m,memory=128Mi

📊 Evidence (1 source(s))

[SignalSource.KUBE_API] SignalKind.PENDING_POD | sp-test-imagepull-84d9d58795-zpfvt | Pod sp-test-imagepull-84d9d58795-zpfvt is in Pending state