About

Imported from jeremylongshore/tons-of-skills-marketplace (skills/.curated/coreweave-common-errors/SKILL.md). Install upstream with npx skills add jeremylongshore/tons-of-skills-marketplace --skill coreweave-common-errors. Copyright stays with the author (MIT).

CoreWeave Common Errors

Name: coreweave-common-errors
Author: jeremylongshore

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

Identify the pod, Job, or Service and collect its status plus recent events.
Match the symptom to the table below, then validate the proposed cause with the listed read-only command before applying a fix.
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

kubectl describe pod <pod-name> | grep -A5 Events
# "0/N nodes are available: insufficient nvidia.com/gpu"

Fix: Check GPU availability: kubectl get nodes -l gpu.nvidia.com/class=A100_PCIE_80GB. Try a different GPU type or region.

2. CUDA Out of Memory

torch.cuda.OutOfMemoryError: 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:

kubectl create secret docker-registry regcred \
  --docker-server=ghcr.io \
  --docker-username=$GH_USER \
  --docker-password=$GH_TOKEN

4. NCCL Timeout (Multi-GPU)

NCCL error: unhandled system error

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:

kubectl get nodes -o json | jq -r '.items[].metadata.labels["gpu.nvidia.com/class"]' | sort -u

7. Service Not Reachable

Fix: Check Service and Endpoints:

kubectl get svc,endpoints <service-name>

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.
Quota or capacity change is required	Obtain the namespace owner approval; do not alter cluster-wide quotas.
Credential failure is suspected	Rotate/revoke through the approved secret manager; never print or paste the credential.
Data or model artifact may be corrupt	Quarantine it and verify checksum before retrying.

Examples

For a Pending GPU pod, collect events and quota before selecting another GPU class:

kubectl -n research describe pod trainer-0
kubectl -n research describe resourcequota
kubectl get nodes -l gpu.nvidia.com/class=A100_PCIE_80GB

If capacity is unavailable, leave the Job unchanged and escalate with the redacted events. Do not remove affinity or quota controls merely to force scheduling.

Resources

Next Steps

For diagnostics, see coreweave-debug-bundle.

Use it

Copy one of these into your project. Installing also returns the manifest and these snippets.

yaml

targets:
  - https://api.opensmartroute.ai/api/v1/registry/jeremylongshore-tons-of-skills-marketplace-coreweave-com-802bcd/manifest   # or paste the manifest below

Manifest

An Open Capability Manifest: the router reads it to know what this does, what it costs and when to pick it.

jeremylongshore-tons-of-skills-marketplace-coreweave-com-802bcd.ocm.jsonjson

{
  "ocm": "1",
  "id": "jeremylongshore-tons-of-skills-marketplace-coreweave-com-802bcd",
  "kind": "skill",
  "name": "coreweave-common-errors",
  "description": "Diagnose and fix CoreWeave GPU scheduling, pod, and networking errors. Use when pods are stuck Pending, GPUs are not allocated, or experiencing CUDA and NCCL errors. Trigger with phrases like \"coreweave error\", \"coreweave pod pending\", \"coreweave gpu not found\", \"coreweave debug\", \"fix coreweave\".",
  "publisher": "jeremylongshore",
  "version": "1.0.0",
  "capabilities": {
    "domains": [
      "coding"
    ],
    "tags": [
      "skill-md",
      "saas",
      "gpu-cloud",
      "kubernetes",
      "inference",
      "coreweave",
      "skills-sh"
    ],
    "languages": [
      "en"
    ]
  },
  "quality_prior": 0.6,
  "examples": [
    "Diagnose and fix CoreWeave GPU scheduling, pod, and networking errors. Use when pods are stuck Pending, GPUs are not allocated, or experiencing CUDA and NCCL errors. Trigger with phrases like \"coreweave error\", \"coreweave pod pending\", \"coreweave gpu not found\", \"coreweave debug\", \"fix coreweave\"."
  ],
  "primary": false,
  "metadata": {
    "source": {
      "provider": "skills.sh",
      "repository": "https://github.com/jeremylongshore/tons-of-skills-marketplace",
      "path": "skills/.curated/coreweave-common-errors/SKILL.md",
      "ref": "HEAD",
      "url": "https://github.com/jeremylongshore/tons-of-skills-marketplace/blob/HEAD/skills/.curated/coreweave-common-errors/SKILL.md",
      "key": "jeremylongshore/tons-of-skills-marketplace/skills/.curated/coreweave-common-errors/SKILL.md"
    },
    "compatibility": "Designed for Claude Code",
    "allowed_tools": [
      "Read,",
      "Bash(kubectl:*),",
      "Grep"
    ],
    "license": "MIT"
  },
  "instructions": "# CoreWeave Common Errors\n\n## Overview\n\nUse this triage guide to classify common GPU, Kubernetes, storage, and connectivity\nfailures before changing capacity or credentials. Capture the smallest redacted\nevidence set and use a reversible fix in the affected namespace.\n\n## Prerequisites\n\n- Read-only access to the affected namespace, pod events, quota, and node labels.\n- The workload name, expected GPU class, and a named service or platform owner.\n\n## Instructions\n\n1. Identify the pod, Job, or Service and collect its status plus recent events.\n2. Match the symptom to the table below, then valida",
  "cost": {
    "context_tokens": 912
  }
}

Fetch it by URL: GET /api/v1/registry/jeremylongshore-tons-of-skills-marketplace-coreweave-com-802bcd/manifest?version=1.0.0

Reviews

Star ratings from people who tried it. One review per account; edit yours any time.

No reviews yet. Install it, try it, and be the first to rate it.