Agent skill · data analytics · jeremylongshore
coreweave-common-errors
Diagnose and fix CoreWeave GPU scheduling, pod, and networking errors.
Why this skill is useful
Provides specific commands and troubleshooting steps for diagnosing and fixing CoreWeave GPU-related issues that the AI wouldn't reliably generate on its own.
What it needs
Requires jq, kubectl installed locally. About 2k tokens when loaded. Last updated 2026-08-07. 2,604 stars on the source repository.
What this skill does
CoreWeave Common Errors Overview Use this triage guide to classify common GPU, Kubernetes, storage, and connectivity failures before changing capacity or credentials. Capture the smallest redacted evidence set and use a reversible fix in the affected namespace. Prerequisites Read-only access to the affected namespace, pod events, quota, and node labels. The workload name, expected GPU class, and a named service or platform owner. Instructions 1. Identify the pod, Job, or Service and collect its status plus recent events. 2. Match the symptom to the table below, then validate the proposed cause with the listed read-only command before applying a fix. 3. Make the smallest namespace-scoped change, verify recovery, and record the redacted event/command outcome in the incident or change record. Community-contributed. Not affiliated with, endorsed by, or sponsored by CoreWeave, Inc. CoreWeave is a registered trademark of CoreWeave, Inc. Error Reference 1. Pod Stuck Pending -- No GPU Available Fix: Check GPU availability: kubectl get nodes -l gpu.nvidia.com/class=A100PCIE80GB. Try a different GPU type or region. 2. CUDA Out of Memory Fix: Reduce batch size, enable gradient checkpointing, or use a larger GPU (A100-80GB instead of 40GB). 3. Image Pull BackOff Fix: Create an imagePullSecret: 4. NCCL Timeout (Multi-GPU) Fix: Ensure all GPUs are on the same node (NVLink). For multi-node, use InfiniBand-connected nodes. 5. PVC Not Mounting Fix: Check storage class availability: kubectl get sc. Use CoreWeave storage classes like shared-hdd-ord1 or shared-ssd-ord1. 6. Node Affinity Mismatch Fix: List valid GPU class labels: 7. Service Not Reachable Fix: Check Service and Endpoints: Output A classified failure with supporting redacted events and a bounded recovery action. A verified recovery result or a clear escalation to the platform owner. Error Handling Triage failure Safe response --- --- Root cause remains unclear Stop speculative changes and collect a redacted debug bundle. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills jeremylongshore/coreweave-common-errors