I/O stalls and aborts due to QEDF driver/firmware interaction on HPE XP arrays

Solution Unverified - Updated -

Issue

  • I/O stalls and aborts due to QEDF driver/firmware interaction on HPE XP arrays.
    • Due to a firmware issue on the QEDF card where larger I/Os are split on the XP array, (which is allowed by SPEC) but the firmware resets the offset mid exchange.
    • The issue is only present in RHEL8, not RHEL7

Environment

  • HPE server using Marvell QEDF QL series HBA
  • HPE XP Storage array where large I/O can be split during an exchange
  • Red Hat Enterprise Linux 8.5 (also exposed in earlier 8.x)
  • File systems built with a 1K block size (In this case was EXT4)

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.

Current Customers and Partners

Log in for full access

Log In

New to Red Hat?

Learn more about Red Hat subscriptions

Using a Red Hat product through a public cloud?

How to access this content