> ## Documentation Index
> Fetch the complete documentation index at: https://docs.coreweave.com/llms.txt
> Use this file to discover all available pages before exploring further.

# How do I interpret GPU health events?

CoreWeave surfaces NVIDIA GPU health signals, including XID counts (driver-reported errors) and thermal data, through the Node Details Grafana dashboard. Not every XID is a hardware fault. Several common codes, including XID 13 and XID 31, are application errors in your workload, while hardware events cause CoreWeave's Node lifecycle automation to drain and triage the affected Node automatically.

To classify a code before opening a ticket, see [What do GPU XID error codes mean and what should I do?](https://docs.coreweave.com/support/cks/articles/what-do-gpu-xid-error-codes-mean). For the full list of codes, see the [NVIDIA XID reference](https://docs.nvidia.com/deploy/xid-errors/latest/index.html). [Contact support](https://docs.coreweave.com/support/contact.md) if a Node is repeatedly draining for the same reason or if your workload is hitting GPU errors that aren't surfaced as drain events.

***

[Nodes](https://docs.coreweave.com/support/cks/tags/nodes.md)[Server Errors](https://docs.coreweave.com/support/cks/tags/server-errors.md)
