Bug 2063923 - [PATCH] Confine numad to systems with actual numa nodes
Summary: [PATCH] Confine numad to systems with actual numa nodes
Keywords:
Status: NEW
Alias: None
Product: Red Hat Enterprise Linux 9
Classification: Red Hat
Component: numad
Version: CentOS Stream
Hardware: x86_64
OS: Linux
unspecified
unspecified
Target Milestone: rc
: ---
Assignee: Lukáš Nykrýn
QA Contact: qe-baseos-daemons
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2022-03-14 16:15 UTC by Pat Riehecky
Modified: 2023-08-15 07:28 UTC (History)
3 users (show)

Fixed In Version:
Doc Type: ---
Doc Text:
Clone Of: 2063922
Environment:
Last Closed:
Type: Bug
Target Upstream Version:
Embargoed:

Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Issue Tracker RHELPLAN-115540 0 None None None 2022-03-14 16:23:57 UTC

Description Pat Riehecky 2022-03-14 16:15:22 UTC 
+++ This bug was initially created as a clone of Bug #2063922 +++

Description of problem:

Currently it is trivial to start numad on systems that do not actually have numa awareness.  This includes a number of virtualization solutions.

Version-Release number of selected component (if applicable):numad-0.5-35.20150602git.fc35.x86_64


How reproducible:
100%

Steps to Reproduce:
1. Fire up a VM without numa nodes
2. start numad
3. it runs but can't do anything causing wasted CPU/Memory

Actual results:

The numad.service starts on systems without numa nodes

Expected results:

The service is able to detect if numa is not present

Additional info:

This patch alters the requirements to ignore starting numad on systems where there are not nodes to migrate between.
Note You need to log in before you can comment on or make changes to this bug.