Azure / az-hop

The Azure HPC On-Demand Platform provides an HPC Cluster Ready solution
https://azure.github.io/az-hop/
MIT License
62 stars 52 forks source link

fix return_to_idle to add dead nodes #1895

Open xpillons opened 3 months ago

xpillons commented 3 months ago

dead nodes are not treated by the return_to_idle.sh script, add the command to mark these nodes down.

scontrol update nodename=$(sinfo --dead | grep idle~ | cut -d'~' -f 2 | xargs | sed 's/ /,/g') state=down reason=dead