Ticket 1297

Summary:	Draining nodes are displayed as mixed in sinfo
Product:	Slurm	Reporter:	David Gloe <david.gloe>
Component:	Other	Assignee:	David Bigagli <david>
Status:	RESOLVED FIXED	QA Contact:
Severity:	4 - Minor Issue
Priority:	---	CC:	brian, da
Version:	14.11.0
Hardware:	Linux
OS:	Linux
Site:	CRAY	Alineos Sites:	---
Atos/Eviden Sites:	---	Confidential Site:	---
Coreweave sites:	---	Cray Sites:	---
DS9 clusters:	---	HPCnow Sites:	---
HPE Sites:	---	IBM Sites:	---
NOAA SIte:	---	OCF Sites:	---
Recursion Pharma Sites:	---	SFW Sites:	---
SNIC sites:	---	Linux Distro:	---
Machine Name:		CLE Version:
Version Fixed:	Slurm 14.11.2	Target Release:	---
DevPrio:	---	Emory-Cloud Sites:	---

Description David Gloe 2014-12-04 05:48:14 MST

Rather than state draining, sinfo shows mixed in 14.11. Is this intended?

dgloe@opal-p2:~> srun sleep 1000&
[1] 12046
dgloe@opal-p2:~> sinfo
PARTITION AVAIL JOB_SIZE  TIMELIMIT   CPUS  S:C:T   NODES STATE      NODELIST
workq*    up    1-infini 1-00:00:00     16  1:8:2       1 mixed      nid00032
workq*    up    1-infini 1-00:00:00    16+ 1+:8+:      23 idle       nid000[20-27,33-35,40-43,48-51,56-59]
nid000[20-27,33-35,40-43,48-51,56-59]
dgloe@opal-p2:~> sudo scontrol update NodeName=nid000[32,33] State=DRAIN Reason='Testing for 820074'
dgloe@opal-p2:~> sinfo
PARTITION AVAIL JOB_SIZE  TIMELIMIT   CPUS  S:C:T   NODES STATE      NODELIST
workq*    up    1-infini 1-00:00:00     16  1:8:2       1 mixed      nid00032
workq*    up    1-infini 1-00:00:00     16  1:8:2       1 drained    nid00033
workq*    up    1-infini 1-00:00:00    16+ 1+:8+:      22 idle       nid000[20-27,34-35,40-43,48-51,56-59]

The node without a job on it correctly shows drained but the one with a job on it is mixed. Even if the job is exclusive, the node is shown as mixed when it is changed to drain.

dgloe@opal-p2:~> srun --exclusive sleep 1000&
[2] 12253
dgloe@opal-p2:~> sinfo
PARTITION AVAIL JOB_SIZE  TIMELIMIT   CPUS  S:C:T   NODES STATE      NODELIST
workq*    up    1-infini 1-00:00:00     16  1:8:2       1 mixed      nid00032
workq*    up    1-infini 1-00:00:00     16  1:8:2       1 drained    nid00033
workq*    up    1-infini 1-00:00:00     32  2:8:2       1 allocated  nid00020
workq*    up    1-infini 1-00:00:00    16+ 1+:8+:      21 idle       nid000[21-27,34-35,40-43,48-51,56-59]
dgloe@opal-p2:~> sudo scontrol update NodeName=nid00020 State=DRAIN Reason='Testing for 820074'
dgloe@opal-p2:~> sinfo
PARTITION AVAIL JOB_SIZE  TIMELIMIT   CPUS  S:C:T   NODES STATE      NODELIST
workq*    up    1-infini 1-00:00:00    16+ 1+:8:2       2 mixed      nid000[20,32]
workq*    up    1-infini 1-00:00:00     16  1:8:2       1 drained    nid00033
workq*    up    1-infini 1-00:00:00    16+ 1+:8+:      21 idle       nid000[21-27,34-35,40-43,48-51,56-59]

Comment 1 David Bigagli 2014-12-04 07:23:17 MST

No, it should be displayed as draining. Commit 0390788c93d.

David