AWS EC2 instances lose access to NVMe volumes due to nvme io_timeout
Issue
- AWS EC2 instances built on the AWS Nitro System can suffer volume access loss when using NVMe EBS volumes, because of a relatively tight io_timeout triggering I/O errors.
- When such volumes are used for root and timeouts happen, the instance can be come unresponsive.
- How set set io_timeout value to prevent timeouts with AWS EC2 nvme storage.
Environment
- AWS Nitro-based instance
- nvme_core.io_timeout set to low values for emulated devices
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.
