CoreWeave Webhooks & Events
Kubernetes Event Monitoring
# Watch GPU pod events
kubectl get events --watch --field-selector=reason=Scheduled,reason=Pulled,reason=Failed
# Monitor GPU utilization via exec
kubectl exec -it deployment/inference -- nvidia-smi --query-gpu=utilization.gpu,memory.used --format=csv -l 5
Prometheus GPU Metrics
# DCGM exporter for GPU metrics (pre-installed on CKS)
# Key metrics:
# DCGM_FI_DEV_GPU_UTIL - GPU utilization %
# DCGM_FI_DEV_FB_USED - GPU memory used
# DCGM_FI_DEV_POWER_USAGE - Power draw
Slack Alert Integration
import subprocess, json, requests
def check_inference_health(deployment: str, slack_url: str):
result = subprocess.run(
["kubectl", "get", "deployment", deployment, "-o", "json"],
capture_output=True, text=True,
)
deploy = json.loads(result.stdout)
ready = deploy["status"].get("readyReplicas", 0)
desired = deploy["spec"]["replicas"]
if ready < desired:
requests.post(slack_url, json={
"text": f"CoreWeave: {deployment} has {ready}/{desired} replicas ready"
})
Resources
Next Steps
For performance optimization, see coreweave-performance-tuning.
1---2name: coreweave-webhooks-events3description: |4 Monitor CoreWeave cluster events and GPU workload status.5 Use when tracking pod lifecycle events, monitoring GPU utilization,6 or alerting on inference service health changes.7 Trigger with phrases like "coreweave events", "coreweave monitoring",8 "coreweave pod alerts", "coreweave gpu monitoring".9allowed-tools: Read, Write, Edit, Bash(kubectl:*)10version: 1.0.011license: MIT12author: Jeremy Longshore <jeremy@intentsolutions.io>13tags: [saas, gpu-cloud, kubernetes, inference, coreweave]14compatible-with: claude-code15---1617# CoreWeave Webhooks & Events1819## Kubernetes Event Monitoring2021```bash22# Watch GPU pod events23kubectl get events --watch --field-selector=reason=Scheduled,reason=Pulled,reason=Failed2425# Monitor GPU utilization via exec26kubectl exec -it deployment/inference -- nvidia-smi --query-gpu=utilization.gpu,memory.used --format=csv -l 527```2829## Prometheus GPU Metrics3031```yaml32# DCGM exporter for GPU metrics (pre-installed on CKS)33# Key metrics:34# DCGM_FI_DEV_GPU_UTIL - GPU utilization %35# DCGM_FI_DEV_FB_USED - GPU memory used36# DCGM_FI_DEV_POWER_USAGE - Power draw37```3839## Slack Alert Integration4041```python42import subprocess, json, requests4344def check_inference_health(deployment: str, slack_url: str):45 result = subprocess.run(46 ["kubectl", "get", "deployment", deployment, "-o", "json"],47 capture_output=True, text=True,48 )49 deploy = json.loads(result.stdout)50 ready = deploy["status"].get("readyReplicas", 0)51 desired = deploy["spec"]["replicas"]5253 if ready < desired:54 requests.post(slack_url, json={55 "text": f"CoreWeave: {deployment} has {ready}/{desired} replicas ready"56 })57```5859## Resources6061- [CoreWeave Observability](https://www.coreweave.com/observability)6263## Next Steps6465For performance optimization, see `coreweave-performance-tuning`.66
jeremylongshore