From adc90f1ca3702ac436f10bffc0eb5cf6e10781d6 Mon Sep 17 00:00:00 2001 From: Ryan Anderson Date: Wed, 12 Jun 2024 09:46:00 -0400 Subject: [PATCH] Remove `cloud_dns` from the `SlurmctldParameters` to avoid the slurm fanout IP address update issue. It is also not required since we update the IP addresses when instances launch --- CHANGELOG.md | 1 + .../templates/default/slurm/slurm.conf.erb | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index 12fb70e5b4..0b6b1a9595 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -17,6 +17,7 @@ This file is used to list changes made in each version of the AWS ParallelCluste **BUG FIXES** - Fix issue with `SharedStorageType: Efs` not bootstrapping on ARM instances. +- Remove `cloud_dns` from the `SlurmctldParameters` in the Slurm config to avoid Slurm fanout issues. This is also not required since we set the IP addresses on instance launch 3.9.1 ------ diff --git a/cookbooks/aws-parallelcluster-slurm/templates/default/slurm/slurm.conf.erb b/cookbooks/aws-parallelcluster-slurm/templates/default/slurm/slurm.conf.erb index 6227c04fea..80455e6f80 100644 --- a/cookbooks/aws-parallelcluster-slurm/templates/default/slurm/slurm.conf.erb +++ b/cookbooks/aws-parallelcluster-slurm/templates/default/slurm/slurm.conf.erb @@ -29,7 +29,7 @@ LaunchParameters=enable_nss_slurm SlurmctldParameters=idle_on_node_suspend,power_save_min_interval=30,node_reg_mem_percent=<%= node['cluster']['slurm_node_reg_mem_percent'] %> TreeWidth=65533 <% else -%> -SlurmctldParameters=idle_on_node_suspend,power_save_min_interval=30,cloud_dns,node_reg_mem_percent=<%= node['cluster']['slurm_node_reg_mem_percent'] %> +SlurmctldParameters=idle_on_node_suspend,power_save_min_interval=30,node_reg_mem_percent=<%= node['cluster']['slurm_node_reg_mem_percent'] %> TreeWidth=30 <% end -%> SuspendProgram=<%= node['cluster']['scripts_dir'] %>/slurm/slurm_suspend