Is the svctm's output of iostat unreliable as described within the man page?

Solution In Progress - Updated -

Issue

  • The following is a portion of iostat's man page. It provides a warning and indicates the svctm field will be removed in a future sysstat version.

    # man iostat 
:
svctm     The average service time(in milliseconds) for I/O requests that were issued to the device.
          Warning! Do not trust this field any more. This field will be removed in a future sysstat version.

  • If the output of svctm from iostat cannot be trusted, are there any alternative fields or tools instead of svctm field?

  • The iostat output shows await much larger than svctm. There is other linux documentation on the web that states that since await time is OS level waiting and the svctm is disk response time, then this means await - svctm is linux kernel overhead and the numbers are showing a kernel bug or need for tuning. Is this correct?

Environment

  • Red Hat Enterprise Linux 9
  • Red Hat Enterprise Linux 8
  • Red Hat Enterprise Linux 7
  • Red Hat Enterprise Linux 6
  • Exceptions:
    • RHEL 9
      • Due to its general unreliable nature, the svctm column has been removed from iostat output starting with RHEL 9. Per systats 🔗 git repository:
        2018/12/14: Version 12.1.2 - Sebastien Godard (sysstat  orange.fr)
* iostat/sar: Remove service time (svctm) metric.
    Service time value (displayed as svctm by sar and iostat) could no
    longer be trusted and has been made obsolete for several years. So
    remove it now from sar (sar -d) and iostat output.

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