You should have all of your nodes defined in the file. For example, here's an excerpt from my test cluster. Every node gets the exact same file.
# COMPUTE NODES NodeName=electro-n1e0 CPUs=8 RealMemory=11264 State=UNKNOWN NodeName=electro-n[5-10] CPUs=8 RealMemory=63488 State=UNKNOWN Feature=avx NodeName=electro-n[11-12] CPUs=8 REalMemory=11264 State=UNKNOWN # PartitionName=normal Nodes=electro-n[5-12],electro-n1e0,electro-n2e0,electro-n3e0,electro-n4e0 Default=YES State=UP On Mon, Feb 2, 2015 at 11:14 AM, Slurm User <slurmu...@gmail.com> wrote: > Hi Bruce > Thanks for your reply. So, how do I handle a situation where "node-1" has > memory of "32GB", and "node-2" has memory of "64GB"? What do I use for > RealMemory (for cons_res)? > Thanks > > On Mon, Feb 2, 2015 at 10:09 AM, Bruce Roberts <schedulerk...@gmail.com> > wrote: > >> Yes. All nodes and their resources need to be defined in the slurm.conf >> on each node, not a different .conf on each node. >> >> >> On 02/02/2015 10:04 AM, Slurm User wrote: >> >> Hello >> >> The documentation says that "slurm.conf" must be consistent across all >> nodes. >> >> How do we handle parameters that are different on a per node basis? >> >> For example, RealMemory may be X on node 1, but Y on node 2. >> >> Am I missing something? >> >> >> > -- Ian Logan Virtualization and Unix Systems Administrator Information and Communication Techologies - New Mexico State University Phone: 575-646-3054 Email: i...@nmsu.edu
