Every 6 months or so, someone's brand new Slurm cluster fails to start because they used a FQDN for their hostname, and didn't know that Slurmctld requires the short name to be resolvable, as well. It's easy enough to tell them how to fix it, but I don't have a good explanation of why they should need to do so.
Any thoughts? Andy -- Andy Riebs [email protected] Hewlett-Packard Enterprise High Performance Computing Software Engineering +1 404 648 9024 My opinions are not necessarily those of HPE
