PowerStore: I/O error observed on NVMe datastore during NDU

摘要: ESXi host running 8.0 U1 using NVMe/FC and NVMe/TCP datastore may experience I/O errors during PowerStore node reboots/failures.

本文适用于 本文不适用于 本文并非针对某种特定的产品。 本文并非包含所有产品版本。
查看其他资源

症状

Symptoms:
  • Hosts connected using NVMe/FC or NVMe/TCP may experience I/O errors during node reboots or NDU.

Environment:
  • ESXi Host 8.0 U1 is connected to PowerStore volumes/vVols. 
  • Host is connected using NVMe/FC or NVMe/TCP.

原因

This is a VMware regression issue introduced in ESXI version 8. 0 U1. In this release, ESXi introduced a Native NVMe stack (aka E2E NVMe stack). Certain error codes that had a retry in previous versions of ESXI no longer have a retry applied. As a result the end applications may experience I/O failures.

解决方案

Upgrade ESXI to version 8.0 U1c or later.

其他信息

VMware KB Article: Rebooting the storage node on the target side may affect VM IOs using the vVOL datastore. (94106)

受影响的产品

PowerStore
文章属性
文章编号: 000216664
文章类型: Solution
上次修改时间: 14 6月 2025
版本:  4
从其他戴尔用户那里查找问题的答案
支持服务
检查您的设备是否在支持服务涵盖的范围内。