[slurm-users] Cancel "reboot ASAP" for a node

Hanby, Mike mhanby at uab.edu
Fri Aug 7 16:23:04 UTC 2020

This is what's in /var/log/slurmctld
Invalid node state transition requested for node c01 from=DRAINING to=CANCEL_REBOOT

So it looks like, for version 18.08 at least, you have to first undrain, then cancel reboot:

scontrol update NodeName="c01" State=undrain Reason="cancelling reboot"
scontrol cancel_reboot c01

From: "Hanby, Mike" <mhanby at uab.edu>
Date: Friday, August 7, 2020 at 11:43 AM
To: Slurm User Community List <slurm-users at lists.schedmd.com>
Subject: Cancel "reboot ASAP" for a node

Howdy, (Slurm 18.08)

We have a bunch of node that we've updated to "scontrol reboot ASAP".

We'd like to cancel a few of those. From the man page, it's suggested that either of the following should work, however both report the same error " slurm_update error: Invalid node state specified":

scontrol cancel_reboot c01
scontrol Update NodeName=c01 State=CANCEL_REBOOT

Here's the 'scontrol show node c01' info for reference:

NodeName=c01 Arch=x86_64 CoresPerSocket=12
   CPUAlloc=7 CPUTot=24 CPULoad=7.04
   NodeAddr=c0115 NodeHostName=c01 Version=18.08
   OS=Linux 3.10.0-1062.9.1.el7.x86_64 #1 SMP Mon Dec 2 08:31:54 EST 2019
   RealMemory=191877 AllocMem=6536 FreeMem=176717 Sockets=2 Boards=1
   State=MIXED+DRAIN ThreadsPerCore=1 TmpDisk=887366 Weight=1 Owner=N/A MCS_label=N/A
   BootTime=2020-07-08T23:16:27 SlurmdStartTime=2020-07-08T23:32:05
   CurrentWatts=0 LowestJoules=0 ConsumedJoules=0
   ExtSensorsJoules=n/s ExtSensorsWatts=0 ExtSensorsTemp=n/s
   Reason=Reboot ASAP [root at 2020-08-06T10:29:22]

Any thoughts as to how to cancel the reboot?

Mike Hanby
mhanby @ uab.edu
Systems Analyst III - Enterprise
IT Research Computing Services
The University of Alabama at Birmingham
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20200807/6b2f0f3e/attachment.htm>

More information about the slurm-users mailing list