Description
GKE workload troubleshooting is an agent skill for diagnosing application workloads that fail on Google Kubernetes Engine. It covers the common failure states, including CrashLoopBackOff, OOMKilled, ImagePullBackOff and Pending pods, and works from logs and events. The skill runs without interaction and keeps a strict read-only boundary: it gathers evidence and proposes fixes, which can be Kubernetes manifest or configuration patches or Google Cloud changes such as IAM bindings or node pool recreation, but it never runs live changes itself. Its first step is context discovery. It extracts the project, cluster, cluster location, workload name and namespace from your request or from the active settings and environment defaults, and it sets a time window for the investigation. The skill is aimed at application developers and on-call engineers who need a structured first pass when a deployment will not start or keeps restarting. It is published by Google on skills.sh under the google skills repository, with about 4.2K installs. The skills.sh listing shows Gen Agent Trust Hub and Socket passing, with Snyk showing Warn.…
Highlights
- Covers CrashLoopBackOff, OOMKilled, ImagePullBackOff and Pending
- Works from logs and events
- Read-only boundary with proposed fixes
- Extracts cluster and workload context automatically
- Sets a time window for the investigation
Ratings & Reviews
No ratings yet
Be the first to rate gke-workload-troubleshooting.
People also visit
gke-cluster-autoscaler
Trigger on mention of GKE cluster autoscaler, node autoscaling
google-cloud-solution-guided-gke-ai-migration
Guides the migration of existing AI workloads
agent-platform-troubleshooting
Troubleshoots Google Cloud Gemini Enterprise Agent Platform issues
gke-workload-troubleshooting FAQ
What failures does it cover?
CrashLoopBackOff, OOMKilled, ImagePullBackOff, Pending pods and other failures seen in logs and events.
Will it change my workload?
No. It proposes manifest, configuration or Google Cloud changes for you to apply.
What context does it need?
The project, cluster, cluster location, workload name and namespace, taken from your request or environment defaults.
How do I install it?
Run npx skills add https://github.com/google/skills --skill gke-workload-troubleshooting.
What are the best gke-workload-troubleshooting alternatives?
Popular alternatives to gke-workload-troubleshooting include gke-node-notready, gke-backup-dr and gke-cluster-autoscaler. See all gke-workload-troubleshooting alternatives