I have tried that and it does work but the reason persists after the nodes get to an idle state.  It's a bit confusing for the node to be idle after a reboot when the reason column still says "rebooting" or "down for maintenance" or whatever.

From: Christopher Samuel via slurm-users <slurm-users@lists.schedmd.com>
Sent: Wednesday, April 15, 2026 12:50 PM
To: slurm-users@lists.schedmd.com <slurm-users@lists.schedmd.com>
Subject: [slurm-users] Re: [Non-DoD Source] Re: Can't clear the REASON
 
On 4/15/26 8:05 am, Berg, Stephen P CIV USN NRL DET SSC MS (USA) via
slurm-users wrote:

> While I've been fiddling with it I've used "test", "testing", "cause"
> and earlier this morning I set all the nodes to "-" just to see if it
> would take that. Just noticed that after a couple hours 91 of the 92
> nodes still have "-" in the REASON column, but one of them now shows up
> as "none".
>
> If the flag is truly not set, or set to NULL does it show up as blank in
> the "sinfo -Nl" output or would it show as none like I'm used to seeing?

[oops - accidentally replied privately - this time to the list!]

Ah - to resume a node you just do:

scontrol update node=$NODE state=resume

Don't try and set the reason field for it.

--
Chris Samuel  :  https://usg01.safelinks.protection.office365.us/?url=http%3A%2F%2Fwww.csamuel.org%2F&data=05%7C02%7Cstephen.p.berg.civ%40us.navy.mil%7Cd10378b58e884c9117f808de9b199f1f%7Ce3333e00c8774b87b6ad45e942de1750%7C0%7C0%7C639118731517219544%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=z6JnVqMrwuXSRAGm9r7Ui30h%2FEluBF2yRxL%2BM0lnyMI%3D&reserved=0  :  Philadelphia, PA, USA

--
slurm-users mailing list -- slurm-users@lists.schedmd.com
To unsubscribe send an email to slurm-users-leave@lists.schedmd.com