[TEMPORARY][B200] Move nvlsm at the beginning of nvidia install to fail fast.

gmarciani · gmarciani · commit 26a710d4648f · 2025-09-11T12:48:53.000-04:00
diff --git a/cookbooks/aws-parallelcluster-platform/recipes/install.rb b/cookbooks/aws-parallelcluster-platform/recipes/install.rb
@@ -17,6 +17,7 @@
 package_repos 'setup the repositories'
 include_recipe "aws-parallelcluster-platform::directories"
 install_packages 'Install OS and extra packages'
+nvidia_nvlsm 'Install Nvidia nvlsm'
 include_recipe "aws-parallelcluster-platform::cookbook_virtualenv"
 include_recipe "aws-parallelcluster-platform::awscli"
 unless alinux2023_on_docker? # Running this recipe on Alinux 2023 docker generates false failure.
diff --git a/cookbooks/aws-parallelcluster-platform/recipes/install/nvidia_install.rb b/cookbooks/aws-parallelcluster-platform/recipes/install/nvidia_install.rb
@@ -27,4 +27,4 @@
 
 nvidia_imex 'Install nvidia-imex'
 
-nvidia_nvlsm 'Install Nvidia nvlsm'
+# nvidia_nvlsm 'Install Nvidia nvlsm'

Original file line number	Diff line number	Diff line change
`@@ -27,4 +27,4 @@`
`27`	`27`
`28`	`28`	`nvidia_imex 'Install nvidia-imex'`
`29`	`29`
`30`		`-nvidia_nvlsm 'Install Nvidia nvlsm'`
	`30`	`+# nvidia_nvlsm 'Install Nvidia nvlsm'`