Skip to content

Commit a58cc62

Browse files
himani2411Himani Anil Deshpande
andauthored
[NVIDIA_IMEX] Change the Imex configuration file names to use queue and compute resource name instead of Launch Template Logical ID (#3024)
Co-authored-by: Himani Anil Deshpande <[email protected]>
1 parent 6c17bff commit a58cc62

File tree

2 files changed

+15
-12
lines changed

2 files changed

+15
-12
lines changed

cookbooks/aws-parallelcluster-platform/resources/nvidia_imex/partial/_nvidia_imex_common.rb

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -92,11 +92,11 @@ def nvidia_enabled_or_installed?
9292
end
9393

9494
def nvidia_imex_main_conf_file
95-
"#{node['cluster']['nvidia']['imex']['shared_dir']}/config_#{node['cluster']['launch_template_id']}.cfg"
95+
"#{node['cluster']['nvidia']['imex']['shared_dir']}/config_#{node['cluster']['scheduler_queue_name']}_#{node['cluster']['scheduler_compute_resource_name']}.cfg"
9696
end
9797

9898
def nvidia_imex_nodes_conf_file
99-
"#{node['cluster']['nvidia']['imex']['shared_dir']}/nodes_config_#{node['cluster']['launch_template_id']}.cfg"
99+
"#{node['cluster']['nvidia']['imex']['shared_dir']}/nodes_config_#{node['cluster']['scheduler_queue_name']}_#{node['cluster']['scheduler_compute_resource_name']}.cfg"
100100
end
101101

102102
def enable_force_configuration?

cookbooks/aws-parallelcluster-platform/spec/unit/resources/nvidia_imex_spec.rb

Lines changed: 13 additions & 10 deletions
Original file line numberDiff line numberDiff line change
@@ -5,7 +5,8 @@
55
nvidia_imex_shared_dir = "SHARED_DIR/nvidia-imex"
66
imex_binary = '/usr/bin/nvidia-imex'
77
imex_ctl_binary = '/usr/bin/nvidia-imex-ctl'
8-
launch_template_id = 'lt-123456789012'
8+
queue_name = 'queue-name'
9+
compute_resource_name = 'compute-resource-name'
910
cluster_artifacts_s3_url = 'https://aws_region-aws-parallelcluster.s3.aws_region.AWS_DOMAIN'
1011

1112
class ConvergeNvidiaImex
@@ -330,50 +331,52 @@ def self.configure(chef_run)
330331
chef_run.node.override['cluster']['nvidia']['imex']['force_configuration'] = force_indicator
331332
chef_run.node.override['cluster']['nvidia']['imex']['shared_dir'] = nvidia_imex_shared_dir
332333
chef_run.node.override['cluster']['node_type'] = node_type
333-
chef_run.node.override['cluster']['launch_template_id'] = launch_template_id
334+
chef_run.node.override['cluster']['scheduler_queue_name'] = queue_name
335+
chef_run.node.override['cluster']['scheduler_compute_resource_name'] = compute_resource_name
336+
334337
ConvergeNvidiaImex.configure(chef_run)
335338
end
336339

337340
if (platform == 'amazon' && version == '2') || %w(HeadNode LoginNode).include?(node_type)
338341
it 'does not configure nvidia-imex' do
339-
is_expected.not_to create_if_missing_template("#{nvidia_imex_shared_dir}/nodes_config_#{launch_template_id}.cfg")
342+
is_expected.not_to create_if_missing_template("#{nvidia_imex_shared_dir}/nodes_config_#{queue_name}_#{compute_resource_name}.cfg")
340343
.with(source: 'nvidia-imex/nvidia-imex-nodes.erb')
341344
.with(user: 'root')
342345
.with(group: 'root')
343346
.with(mode: '0755')
344-
is_expected.not_to create_if_missing_template("#{nvidia_imex_shared_dir}/config_#{launch_template_id}.cfg")
347+
is_expected.not_to create_if_missing_template("#{nvidia_imex_shared_dir}/config_#{queue_name}_#{compute_resource_name}.cfg")
345348
.with(source: 'nvidia-imex/nvidia-imex-config.erb')
346349
.with(user: 'root')
347350
.with(group: 'root')
348351
.with(mode: '0755')
349-
.with(variables: { imex_nodes_config_file_path: "#{nvidia_imex_shared_dir}/nodes_config_#{launch_template_id}.cfg" })
352+
.with(variables: { imex_nodes_config_file_path: "#{nvidia_imex_shared_dir}/nodes_config_#{queue_name}_#{compute_resource_name}.cfg" })
350353
is_expected.not_to create_template("/etc/systemd/system/nvidia-imex.service")
351354
.with(source: 'nvidia-imex/nvidia-imex.service.erb')
352355
.with(user: 'root')
353356
.with(group: 'root')
354357
.with(mode: '0644')
355-
.with(variables: { imex_main_config_file_path: "#{nvidia_imex_shared_dir}/config_#{launch_template_id}.cfg" })
358+
.with(variables: { imex_main_config_file_path: "#{nvidia_imex_shared_dir}/config_#{queue_name}_#{compute_resource_name}.cfg" })
356359
is_expected.not_to start_service('nvidia-imex').with_action(%i(enable start)).with_supports({ status: true })
357360
end
358361
else
359362
it 'it starts nvidia-imex service' do
360-
is_expected.to create_if_missing_template("#{nvidia_imex_shared_dir}/nodes_config_#{launch_template_id}.cfg")
363+
is_expected.to create_if_missing_template("#{nvidia_imex_shared_dir}/nodes_config_#{queue_name}_#{compute_resource_name}.cfg")
361364
.with(source: 'nvidia-imex/nvidia-imex-nodes.erb')
362365
.with(user: 'root')
363366
.with(group: 'root')
364367
.with(mode: '0755')
365-
is_expected.to create_if_missing_template("#{nvidia_imex_shared_dir}/config_#{launch_template_id}.cfg")
368+
is_expected.to create_if_missing_template("#{nvidia_imex_shared_dir}/config_#{queue_name}_#{compute_resource_name}.cfg")
366369
.with(source: 'nvidia-imex/nvidia-imex-config.erb')
367370
.with(user: 'root')
368371
.with(group: 'root')
369372
.with(mode: '0755')
370-
.with(variables: { imex_nodes_config_file_path: "#{nvidia_imex_shared_dir}/nodes_config_#{launch_template_id}.cfg" })
373+
.with(variables: { imex_nodes_config_file_path: "#{nvidia_imex_shared_dir}/nodes_config_#{queue_name}_#{compute_resource_name}.cfg" })
371374
is_expected.to create_template("/etc/systemd/system/nvidia-imex.service")
372375
.with(source: 'nvidia-imex/nvidia-imex.service.erb')
373376
.with(user: 'root')
374377
.with(group: 'root')
375378
.with(mode: '0644')
376-
.with(variables: { imex_main_config_file_path: "#{nvidia_imex_shared_dir}/config_#{launch_template_id}.cfg" })
379+
.with(variables: { imex_main_config_file_path: "#{nvidia_imex_shared_dir}/config_#{queue_name}_#{compute_resource_name}.cfg" })
377380
is_expected.to start_service('nvidia-imex').with_action(%i(enable start)).with_supports({ status: true })
378381
end
379382
end

0 commit comments

Comments
 (0)