Ticket 5348

Summary:	Suspending Nodes which are not in IDLE mode
Product:	Slurm	Reporter:	Robert Klemm <roklemm>
Component:	slurmctld	Assignee:	Jacob Jenson <jacob>
Status:	RESOLVED INVALID	QA Contact:
Severity:	6 - No support contract
Priority:	---	CC:	fzillner
Version:	17.11.7
Hardware:	Linux
OS:	Linux
Site:	-Other-	Slinky Site:	---
Alineos Sites:	---	Atos/Eviden Sites:	---
Confidential Site:	---	Coreweave sites:	---
Cray Sites:	---	DS9 clusters:	---
Google sites:	---	HPCnow Sites:	---
HPE Sites:	---	IBM Sites:	---
NOAA SIte:	---	NoveTech Sites:	---
Nvidia HWinf-CS Sites:	---	OCF Sites:	---
Recursion Pharma Sites:	---	SFW Sites:	---
SNIC sites:	---	Tzag Elita Sites:	---
Linux Distro:	---	Machine Name:
CLE Version:		Version Fixed:
Target Release:	---	DevPrio:	---
Emory-Cloud Sites:	---

Description Robert Klemm 2018-06-25 03:32:00 MDT

Hey folks,

when using PowerSaving, nodes which state is not IDLE, but DOWN, DRAIN, FAIL or even in MAINT, get set into PowerSaving-Mode and will be suspended.

Comment 1 fzillner 2020-05-15 07:28:51 MDT

I'm having the same problem. OpenHPC slurm 18.08.8 on CentOS 7.7.

The only workaround I can think of right now is to double check in my suspend script if the node is really IDLE or not. But IMO slurm should only suspend the nodes that are idle or down, at least that's what's stated in the docs.