aws-sdk-pcs 1.67.0 → 1.69.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
checksums.yaml CHANGED
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  SHA256:
3
- metadata.gz: 69196d4c06fc93e691f3a6ea557ddd03fdf60dbeb6f96c51fa0cb943f2a596ce
4
- data.tar.gz: c6d84870ae2cf1995abe10c3c230e96efd3f7804b021101693787d587ee33ec9
3
+ metadata.gz: 05c343f6c8b1b99ae6140ca736cc5a1d492ac23beaf49bc4a5fa1929c1679a19
4
+ data.tar.gz: b7653dc51c7fc4cf70568596cbe0813b34c9455164ac36c46ee717c03a9f0e45
5
5
  SHA512:
6
- metadata.gz: 7f7c104c059c5b475cab3fceb2bc97fa765b1284293ee3963518e7594c9c22dfdedfc1e76c1c95ef110d33f33b69fbefeb8c73c2bfb736158cfc7115bb73b781
7
- data.tar.gz: a89e2641a0955d5a8b5cc4002045db2d6e0ac31c2ca2ba9f211b65468f88f65aaedc4e4630da2b5d1463b2fb51ae5cc525e48a4bb603443d7806573073db3bfd
6
+ metadata.gz: cad8bb3939f93006fa37fec035c413b7f390721900e09a7ea4e40571ade21eea5224cb6b64e8c44ea11e1da61d92d2970691ace0412cccd3ad2ba88411adbd07
7
+ data.tar.gz: 4b24a46df91e3792bdc89dd8aa2fd23536393cb873b8310e611798f41ced1316447c9d576da0e6b3a19af8d9a5c25aaeffdf4d696c01fbea36626e359938ec4c
data/CHANGELOG.md CHANGED
@@ -1,6 +1,16 @@
1
1
  Unreleased Changes
2
2
  ------------------
3
3
 
4
+ 1.69.0 (2026-11-20)
5
+ ------------------
6
+
7
+ * Feature - Added support for Instance Health Checks in AWS PCS.
8
+
9
+ 1.68.0 (2026-11-19)
10
+ ------------------
11
+
12
+ * Feature - Added support for Job Submit Plugins in AWS PCS.
13
+
4
14
  1.67.0 (2026-11-18)
5
15
  ------------------
6
16
 
data/VERSION CHANGED
@@ -1 +1 @@
1
- 1.67.0
1
+ 1.69.0
@@ -581,6 +581,17 @@ module Aws::PCS
581
581
  # parameter_value: "String", # required
582
582
  # },
583
583
  # ],
584
+ # job_submit_plugins: [
585
+ # {
586
+ # name: "JobSubmitPluginNameString", # required
587
+ # type: "STARLARK", # required, accepts STARLARK
588
+ # script_source: { # required
589
+ # script_location: "JobSubmitScriptSourceScriptLocationString", # required
590
+ # s3_version_id: "JobSubmitScriptSourceS3VersionIdString",
591
+ # checksum: "JobSubmitScriptSourceChecksumString",
592
+ # },
593
+ # },
594
+ # ],
584
595
  # accounting: {
585
596
  # default_purge_time_in_days: 1,
586
597
  # mode: "STANDARD", # required, accepts STANDARD, NONE
@@ -616,6 +627,12 @@ module Aws::PCS
616
627
  # resp.cluster.slurm_configuration.cgroup_custom_settings #=> Array
617
628
  # resp.cluster.slurm_configuration.cgroup_custom_settings[0].parameter_name #=> String
618
629
  # resp.cluster.slurm_configuration.cgroup_custom_settings[0].parameter_value #=> String
630
+ # resp.cluster.slurm_configuration.job_submit_plugins #=> Array
631
+ # resp.cluster.slurm_configuration.job_submit_plugins[0].name #=> String
632
+ # resp.cluster.slurm_configuration.job_submit_plugins[0].type #=> String, one of "STARLARK"
633
+ # resp.cluster.slurm_configuration.job_submit_plugins[0].script_source.script_location #=> String
634
+ # resp.cluster.slurm_configuration.job_submit_plugins[0].script_source.s3_version_id #=> String
635
+ # resp.cluster.slurm_configuration.job_submit_plugins[0].script_source.checksum #=> String
619
636
  # resp.cluster.slurm_configuration.auth_key.secret_arn #=> String
620
637
  # resp.cluster.slurm_configuration.auth_key.secret_version #=> String
621
638
  # resp.cluster.slurm_configuration.jwt_auth.jwt_key.secret_arn #=> String
@@ -782,6 +799,14 @@ module Aws::PCS
782
799
  # of a compute node's lifecycle, such as when a compute node finishes
783
800
  # bootstrapping or becomes ready to accept jobs.
784
801
  #
802
+ # @option params [Types::InstanceHealthChecksRequest] :instance_health_checks
803
+ # The instance health check settings for the compute node group.
804
+ # Instance health checks detect faults on instances in the compute node
805
+ # group and respond with the repair action that you choose. You can
806
+ # configure health monitoring, deep health checks, or both. You can't
807
+ # specify an empty object (`{}`). Instance health checks require Slurm
808
+ # version 26.05 or later.
809
+ #
785
810
  # @option params [String] :client_token
786
811
  # A unique, case-sensitive identifier that you provide to ensure the
787
812
  # idempotency of the request. Idempotency ensures that an API request
@@ -875,6 +900,18 @@ module Aws::PCS
875
900
  # },
876
901
  # script_caching_policy: "CACHE_ONCE", # accepts CACHE_ONCE, REFRESH_ON_REBOOT
877
902
  # },
903
+ # instance_health_checks: {
904
+ # health_monitoring: {
905
+ # mode: "NONE", # accepts NONE, STANDARD
906
+ # repair_action: "AUTO", # accepts AUTO, REPLACE, ISOLATE, NONE
907
+ # },
908
+ # deep_health_checks: {
909
+ # on_start: {
910
+ # profile: "NONE", # accepts NONE, QUICK, MEDIUM, LONG
911
+ # repair_action: "AUTO", # accepts AUTO, REPLACE, ISOLATE, NONE
912
+ # },
913
+ # },
914
+ # },
878
915
  # client_token: "SBClientToken",
879
916
  # tags: {
880
917
  # "TagKey" => "TagValue",
@@ -932,6 +969,10 @@ module Aws::PCS
932
969
  # resp.compute_node_group.node_lifecycle_actions.stages.node_ready[0].on_error #=> String, one of "TERMINATE", "STOP_SEQUENCE", "CONTINUE"
933
970
  # resp.compute_node_group.node_lifecycle_actions.stages.node_ready[0].execution_policy #=> String, one of "FIRST_BOOT_ONLY", "EVERY_BOOT"
934
971
  # resp.compute_node_group.node_lifecycle_actions.script_caching_policy #=> String, one of "CACHE_ONCE", "REFRESH_ON_REBOOT"
972
+ # resp.compute_node_group.instance_health_checks.health_monitoring.mode #=> String, one of "NONE", "STANDARD"
973
+ # resp.compute_node_group.instance_health_checks.health_monitoring.repair_action #=> String, one of "AUTO", "REPLACE", "ISOLATE", "NONE"
974
+ # resp.compute_node_group.instance_health_checks.deep_health_checks.on_start.profile #=> String, one of "NONE", "QUICK", "MEDIUM", "LONG"
975
+ # resp.compute_node_group.instance_health_checks.deep_health_checks.on_start.repair_action #=> String, one of "AUTO", "REPLACE", "ISOLATE", "NONE"
935
976
  # resp.compute_node_group.error_info #=> Array
936
977
  # resp.compute_node_group.error_info[0].code #=> String
937
978
  # resp.compute_node_group.error_info[0].message #=> String
@@ -1189,6 +1230,12 @@ module Aws::PCS
1189
1230
  # resp.cluster.slurm_configuration.cgroup_custom_settings #=> Array
1190
1231
  # resp.cluster.slurm_configuration.cgroup_custom_settings[0].parameter_name #=> String
1191
1232
  # resp.cluster.slurm_configuration.cgroup_custom_settings[0].parameter_value #=> String
1233
+ # resp.cluster.slurm_configuration.job_submit_plugins #=> Array
1234
+ # resp.cluster.slurm_configuration.job_submit_plugins[0].name #=> String
1235
+ # resp.cluster.slurm_configuration.job_submit_plugins[0].type #=> String, one of "STARLARK"
1236
+ # resp.cluster.slurm_configuration.job_submit_plugins[0].script_source.script_location #=> String
1237
+ # resp.cluster.slurm_configuration.job_submit_plugins[0].script_source.s3_version_id #=> String
1238
+ # resp.cluster.slurm_configuration.job_submit_plugins[0].script_source.checksum #=> String
1192
1239
  # resp.cluster.slurm_configuration.auth_key.secret_arn #=> String
1193
1240
  # resp.cluster.slurm_configuration.auth_key.secret_version #=> String
1194
1241
  # resp.cluster.slurm_configuration.jwt_auth.jwt_key.secret_arn #=> String
@@ -1292,6 +1339,10 @@ module Aws::PCS
1292
1339
  # resp.compute_node_group.node_lifecycle_actions.stages.node_ready[0].on_error #=> String, one of "TERMINATE", "STOP_SEQUENCE", "CONTINUE"
1293
1340
  # resp.compute_node_group.node_lifecycle_actions.stages.node_ready[0].execution_policy #=> String, one of "FIRST_BOOT_ONLY", "EVERY_BOOT"
1294
1341
  # resp.compute_node_group.node_lifecycle_actions.script_caching_policy #=> String, one of "CACHE_ONCE", "REFRESH_ON_REBOOT"
1342
+ # resp.compute_node_group.instance_health_checks.health_monitoring.mode #=> String, one of "NONE", "STANDARD"
1343
+ # resp.compute_node_group.instance_health_checks.health_monitoring.repair_action #=> String, one of "AUTO", "REPLACE", "ISOLATE", "NONE"
1344
+ # resp.compute_node_group.instance_health_checks.deep_health_checks.on_start.profile #=> String, one of "NONE", "QUICK", "MEDIUM", "LONG"
1345
+ # resp.compute_node_group.instance_health_checks.deep_health_checks.on_start.repair_action #=> String, one of "AUTO", "REPLACE", "ISOLATE", "NONE"
1295
1346
  # resp.compute_node_group.error_info #=> Array
1296
1347
  # resp.compute_node_group.error_info[0].code #=> String
1297
1348
  # resp.compute_node_group.error_info[0].message #=> String
@@ -1563,6 +1614,7 @@ module Aws::PCS
1563
1614
  # * {Types::RegisterComputeNodeGroupInstanceResponse#compute_node_group_name #compute_node_group_name} => String
1564
1615
  # * {Types::RegisterComputeNodeGroupInstanceResponse#node_lifecycle_actions #node_lifecycle_actions} => Types::NodeLifecycleActions
1565
1616
  # * {Types::RegisterComputeNodeGroupInstanceResponse#scheduler_version #scheduler_version} => String
1617
+ # * {Types::RegisterComputeNodeGroupInstanceResponse#instance_health_checks #instance_health_checks} => Types::InstanceHealthChecks
1566
1618
  #
1567
1619
  # @example Request syntax with placeholder values
1568
1620
  #
@@ -1604,6 +1656,10 @@ module Aws::PCS
1604
1656
  # resp.node_lifecycle_actions.stages.node_ready[0].execution_policy #=> String, one of "FIRST_BOOT_ONLY", "EVERY_BOOT"
1605
1657
  # resp.node_lifecycle_actions.script_caching_policy #=> String, one of "CACHE_ONCE", "REFRESH_ON_REBOOT"
1606
1658
  # resp.scheduler_version #=> String
1659
+ # resp.instance_health_checks.health_monitoring.mode #=> String, one of "NONE", "STANDARD"
1660
+ # resp.instance_health_checks.health_monitoring.repair_action #=> String, one of "AUTO", "REPLACE", "ISOLATE", "NONE"
1661
+ # resp.instance_health_checks.deep_health_checks.on_start.profile #=> String, one of "NONE", "QUICK", "MEDIUM", "LONG"
1662
+ # resp.instance_health_checks.deep_health_checks.on_start.repair_action #=> String, one of "AUTO", "REPLACE", "ISOLATE", "NONE"
1607
1663
  #
1608
1664
  # @see http://docs.aws.amazon.com/goto/WebAPI/pcs-2023-02-10/RegisterComputeNodeGroupInstance AWS API Documentation
1609
1665
  #
@@ -1747,6 +1803,17 @@ module Aws::PCS
1747
1803
  # parameter_value: "String", # required
1748
1804
  # },
1749
1805
  # ],
1806
+ # job_submit_plugins: [
1807
+ # {
1808
+ # name: "JobSubmitPluginNameString", # required
1809
+ # type: "STARLARK", # required, accepts STARLARK
1810
+ # script_source: { # required
1811
+ # script_location: "JobSubmitScriptSourceScriptLocationString", # required
1812
+ # s3_version_id: "JobSubmitScriptSourceS3VersionIdString",
1813
+ # checksum: "JobSubmitScriptSourceChecksumString",
1814
+ # },
1815
+ # },
1816
+ # ],
1750
1817
  # accounting: {
1751
1818
  # default_purge_time_in_days: 1,
1752
1819
  # mode: "STANDARD", # accepts STANDARD, NONE
@@ -1781,6 +1848,12 @@ module Aws::PCS
1781
1848
  # resp.cluster.slurm_configuration.cgroup_custom_settings #=> Array
1782
1849
  # resp.cluster.slurm_configuration.cgroup_custom_settings[0].parameter_name #=> String
1783
1850
  # resp.cluster.slurm_configuration.cgroup_custom_settings[0].parameter_value #=> String
1851
+ # resp.cluster.slurm_configuration.job_submit_plugins #=> Array
1852
+ # resp.cluster.slurm_configuration.job_submit_plugins[0].name #=> String
1853
+ # resp.cluster.slurm_configuration.job_submit_plugins[0].type #=> String, one of "STARLARK"
1854
+ # resp.cluster.slurm_configuration.job_submit_plugins[0].script_source.script_location #=> String
1855
+ # resp.cluster.slurm_configuration.job_submit_plugins[0].script_source.s3_version_id #=> String
1856
+ # resp.cluster.slurm_configuration.job_submit_plugins[0].script_source.checksum #=> String
1784
1857
  # resp.cluster.slurm_configuration.auth_key.secret_arn #=> String
1785
1858
  # resp.cluster.slurm_configuration.auth_key.secret_version #=> String
1786
1859
  # resp.cluster.slurm_configuration.jwt_auth.jwt_key.secret_arn #=> String
@@ -1941,6 +2014,23 @@ module Aws::PCS
1941
2014
  # of a compute node's lifecycle, such as when a compute node finishes
1942
2015
  # bootstrapping or becomes ready to accept jobs.
1943
2016
  #
2017
+ # @option params [Types::UpdateInstanceHealthChecksRequest] :instance_health_checks
2018
+ # The instance health check settings for the compute node group.
2019
+ # Instance health checks detect faults on instances in the compute node
2020
+ # group and respond with the repair action that you choose. You can
2021
+ # configure health monitoring, deep health checks, or both. You can't
2022
+ # specify an empty object (`{}`). Instance health checks require Slurm
2023
+ # version 26.05 or later.
2024
+ #
2025
+ # The settings that you specify replace all of the current instance
2026
+ # health check settings of the compute node group. Include every setting
2027
+ # that you want to keep. If you omit this member, the current settings
2028
+ # don't change. If you omit `healthMonitoring` or `deepHealthChecks`,
2029
+ # that check is turned off. You can also turn off a check: set
2030
+ # `healthMonitoring.mode` or `deepHealthChecks.onStart.profile` to
2031
+ # `NONE` and omit its `repairAction`. When you change these settings,
2032
+ # PCS drains and replaces the instances in the compute node group.
2033
+ #
1944
2034
  # @option params [String] :client_token
1945
2035
  # A unique, case-sensitive identifier that you provide to ensure the
1946
2036
  # idempotency of the request. Idempotency ensures that an API request
@@ -2025,6 +2115,18 @@ module Aws::PCS
2025
2115
  # },
2026
2116
  # script_caching_policy: "CACHE_ONCE", # accepts CACHE_ONCE, REFRESH_ON_REBOOT
2027
2117
  # },
2118
+ # instance_health_checks: {
2119
+ # health_monitoring: {
2120
+ # mode: "NONE", # accepts NONE, STANDARD
2121
+ # repair_action: "AUTO", # accepts AUTO, REPLACE, ISOLATE, NONE
2122
+ # },
2123
+ # deep_health_checks: {
2124
+ # on_start: {
2125
+ # profile: "NONE", # accepts NONE, QUICK, MEDIUM, LONG
2126
+ # repair_action: "AUTO", # accepts AUTO, REPLACE, ISOLATE, NONE
2127
+ # },
2128
+ # },
2129
+ # },
2028
2130
  # client_token: "SBClientToken",
2029
2131
  # })
2030
2132
  #
@@ -2079,6 +2181,10 @@ module Aws::PCS
2079
2181
  # resp.compute_node_group.node_lifecycle_actions.stages.node_ready[0].on_error #=> String, one of "TERMINATE", "STOP_SEQUENCE", "CONTINUE"
2080
2182
  # resp.compute_node_group.node_lifecycle_actions.stages.node_ready[0].execution_policy #=> String, one of "FIRST_BOOT_ONLY", "EVERY_BOOT"
2081
2183
  # resp.compute_node_group.node_lifecycle_actions.script_caching_policy #=> String, one of "CACHE_ONCE", "REFRESH_ON_REBOOT"
2184
+ # resp.compute_node_group.instance_health_checks.health_monitoring.mode #=> String, one of "NONE", "STANDARD"
2185
+ # resp.compute_node_group.instance_health_checks.health_monitoring.repair_action #=> String, one of "AUTO", "REPLACE", "ISOLATE", "NONE"
2186
+ # resp.compute_node_group.instance_health_checks.deep_health_checks.on_start.profile #=> String, one of "NONE", "QUICK", "MEDIUM", "LONG"
2187
+ # resp.compute_node_group.instance_health_checks.deep_health_checks.on_start.repair_action #=> String, one of "AUTO", "REPLACE", "ISOLATE", "NONE"
2082
2188
  # resp.compute_node_group.error_info #=> Array
2083
2189
  # resp.compute_node_group.error_info[0].code #=> String
2084
2190
  # resp.compute_node_group.error_info[0].message #=> String
@@ -2190,7 +2296,7 @@ module Aws::PCS
2190
2296
  tracer: tracer
2191
2297
  )
2192
2298
  context[:gem_name] = 'aws-sdk-pcs'
2193
- context[:gem_version] = '1.67.0'
2299
+ context[:gem_version] = '1.69.0'
2194
2300
  Seahorse::Client::Request.new(handlers, context)
2195
2301
  end
2196
2302
 
@@ -56,6 +56,10 @@ module Aws::PCS
56
56
  CreateQueueRequest = Shapes::StructureShape.new(name: 'CreateQueueRequest')
57
57
  CreateQueueResponse = Shapes::StructureShape.new(name: 'CreateQueueResponse')
58
58
  CustomLaunchTemplate = Shapes::StructureShape.new(name: 'CustomLaunchTemplate')
59
+ DeepHealthCheckOnStart = Shapes::StructureShape.new(name: 'DeepHealthCheckOnStart')
60
+ DeepHealthCheckOnStartProfile = Shapes::StringShape.new(name: 'DeepHealthCheckOnStartProfile')
61
+ DeepHealthCheckOnStartRepairAction = Shapes::StringShape.new(name: 'DeepHealthCheckOnStartRepairAction')
62
+ DeepHealthChecks = Shapes::StructureShape.new(name: 'DeepHealthChecks')
59
63
  DeleteClusterRequest = Shapes::StructureShape.new(name: 'DeleteClusterRequest')
60
64
  DeleteClusterResponse = Shapes::StructureShape.new(name: 'DeleteClusterResponse')
61
65
  DeleteComputeNodeGroupRequest = Shapes::StructureShape.new(name: 'DeleteComputeNodeGroupRequest')
@@ -76,11 +80,24 @@ module Aws::PCS
76
80
  GetQueueResponse = Shapes::StructureShape.new(name: 'GetQueueResponse')
77
81
  GresCustomSettingMap = Shapes::MapShape.new(name: 'GresCustomSettingMap')
78
82
  GresCustomSettings = Shapes::ListShape.new(name: 'GresCustomSettings')
83
+ HealthMonitoring = Shapes::StructureShape.new(name: 'HealthMonitoring')
84
+ HealthMonitoringMode = Shapes::StringShape.new(name: 'HealthMonitoringMode')
85
+ HealthMonitoringRepairAction = Shapes::StringShape.new(name: 'HealthMonitoringRepairAction')
79
86
  InstanceConfig = Shapes::StructureShape.new(name: 'InstanceConfig')
87
+ InstanceHealthChecks = Shapes::StructureShape.new(name: 'InstanceHealthChecks')
88
+ InstanceHealthChecksRequest = Shapes::StructureShape.new(name: 'InstanceHealthChecksRequest')
80
89
  InstanceList = Shapes::ListShape.new(name: 'InstanceList')
81
90
  InstanceProfileArn = Shapes::StringShape.new(name: 'InstanceProfileArn')
82
91
  Integer = Shapes::IntegerShape.new(name: 'Integer')
83
92
  InternalServerException = Shapes::StructureShape.new(name: 'InternalServerException')
93
+ JobSubmitPlugin = Shapes::StructureShape.new(name: 'JobSubmitPlugin')
94
+ JobSubmitPluginList = Shapes::ListShape.new(name: 'JobSubmitPluginList')
95
+ JobSubmitPluginNameString = Shapes::StringShape.new(name: 'JobSubmitPluginNameString')
96
+ JobSubmitPluginType = Shapes::StringShape.new(name: 'JobSubmitPluginType')
97
+ JobSubmitScriptSource = Shapes::StructureShape.new(name: 'JobSubmitScriptSource')
98
+ JobSubmitScriptSourceChecksumString = Shapes::StringShape.new(name: 'JobSubmitScriptSourceChecksumString')
99
+ JobSubmitScriptSourceS3VersionIdString = Shapes::StringShape.new(name: 'JobSubmitScriptSourceS3VersionIdString')
100
+ JobSubmitScriptSourceScriptLocationString = Shapes::StringShape.new(name: 'JobSubmitScriptSourceScriptLocationString')
84
101
  JwtAuth = Shapes::StructureShape.new(name: 'JwtAuth')
85
102
  JwtKey = Shapes::StructureShape.new(name: 'JwtKey')
86
103
  KeyPairName = Shapes::StringShape.new(name: 'KeyPairName')
@@ -175,6 +192,7 @@ module Aws::PCS
175
192
  UpdateComputeNodeGroupResponse = Shapes::StructureShape.new(name: 'UpdateComputeNodeGroupResponse')
176
193
  UpdateComputeNodeGroupSlurmConfigurationRequest = Shapes::StructureShape.new(name: 'UpdateComputeNodeGroupSlurmConfigurationRequest')
177
194
  UpdateComputeNodeGroupSlurmConfigurationRequestScaleDownIdleTimeInSecondsInteger = Shapes::IntegerShape.new(name: 'UpdateComputeNodeGroupSlurmConfigurationRequestScaleDownIdleTimeInSecondsInteger')
195
+ UpdateInstanceHealthChecksRequest = Shapes::StructureShape.new(name: 'UpdateInstanceHealthChecksRequest')
178
196
  UpdateNodeLifecycleActionsRequest = Shapes::StructureShape.new(name: 'UpdateNodeLifecycleActionsRequest')
179
197
  UpdateQueueRequest = Shapes::StructureShape.new(name: 'UpdateQueueRequest')
180
198
  UpdateQueueResponse = Shapes::StructureShape.new(name: 'UpdateQueueResponse')
@@ -223,6 +241,7 @@ module Aws::PCS
223
241
  ClusterSlurmConfiguration.add_member(:slurm_custom_settings, Shapes::ShapeRef.new(shape: SlurmCustomSettings, location_name: "slurmCustomSettings"))
224
242
  ClusterSlurmConfiguration.add_member(:slurmdbd_custom_settings, Shapes::ShapeRef.new(shape: SlurmdbdCustomSettings, location_name: "slurmdbdCustomSettings"))
225
243
  ClusterSlurmConfiguration.add_member(:cgroup_custom_settings, Shapes::ShapeRef.new(shape: CgroupCustomSettings, location_name: "cgroupCustomSettings"))
244
+ ClusterSlurmConfiguration.add_member(:job_submit_plugins, Shapes::ShapeRef.new(shape: JobSubmitPluginList, location_name: "jobSubmitPlugins"))
226
245
  ClusterSlurmConfiguration.add_member(:auth_key, Shapes::ShapeRef.new(shape: SlurmAuthKey, location_name: "authKey"))
227
246
  ClusterSlurmConfiguration.add_member(:jwt_auth, Shapes::ShapeRef.new(shape: JwtAuth, location_name: "jwtAuth"))
228
247
  ClusterSlurmConfiguration.add_member(:accounting, Shapes::ShapeRef.new(shape: Accounting, location_name: "accounting"))
@@ -233,6 +252,7 @@ module Aws::PCS
233
252
  ClusterSlurmConfigurationRequest.add_member(:slurm_custom_settings, Shapes::ShapeRef.new(shape: SlurmCustomSettings, location_name: "slurmCustomSettings"))
234
253
  ClusterSlurmConfigurationRequest.add_member(:slurmdbd_custom_settings, Shapes::ShapeRef.new(shape: SlurmdbdCustomSettings, location_name: "slurmdbdCustomSettings"))
235
254
  ClusterSlurmConfigurationRequest.add_member(:cgroup_custom_settings, Shapes::ShapeRef.new(shape: CgroupCustomSettings, location_name: "cgroupCustomSettings"))
255
+ ClusterSlurmConfigurationRequest.add_member(:job_submit_plugins, Shapes::ShapeRef.new(shape: JobSubmitPluginList, location_name: "jobSubmitPlugins"))
236
256
  ClusterSlurmConfigurationRequest.add_member(:accounting, Shapes::ShapeRef.new(shape: AccountingRequest, location_name: "accounting"))
237
257
  ClusterSlurmConfigurationRequest.add_member(:slurm_rest, Shapes::ShapeRef.new(shape: SlurmRestRequest, location_name: "slurmRest"))
238
258
  ClusterSlurmConfigurationRequest.struct_class = Types::ClusterSlurmConfigurationRequest
@@ -265,6 +285,7 @@ module Aws::PCS
265
285
  ComputeNodeGroup.add_member(:spot_options, Shapes::ShapeRef.new(shape: SpotOptions, location_name: "spotOptions"))
266
286
  ComputeNodeGroup.add_member(:slurm_configuration, Shapes::ShapeRef.new(shape: ComputeNodeGroupSlurmConfiguration, location_name: "slurmConfiguration"))
267
287
  ComputeNodeGroup.add_member(:node_lifecycle_actions, Shapes::ShapeRef.new(shape: NodeLifecycleActions, location_name: "nodeLifecycleActions"))
288
+ ComputeNodeGroup.add_member(:instance_health_checks, Shapes::ShapeRef.new(shape: InstanceHealthChecks, location_name: "instanceHealthChecks"))
268
289
  ComputeNodeGroup.add_member(:error_info, Shapes::ShapeRef.new(shape: ErrorInfoList, location_name: "errorInfo"))
269
290
  ComputeNodeGroup.struct_class = Types::ComputeNodeGroup
270
291
 
@@ -326,6 +347,7 @@ module Aws::PCS
326
347
  CreateComputeNodeGroupRequest.add_member(:spot_options, Shapes::ShapeRef.new(shape: SpotOptions, location_name: "spotOptions"))
327
348
  CreateComputeNodeGroupRequest.add_member(:slurm_configuration, Shapes::ShapeRef.new(shape: ComputeNodeGroupSlurmConfigurationRequest, location_name: "slurmConfiguration"))
328
349
  CreateComputeNodeGroupRequest.add_member(:node_lifecycle_actions, Shapes::ShapeRef.new(shape: NodeLifecycleActionsRequest, location_name: "nodeLifecycleActions"))
350
+ CreateComputeNodeGroupRequest.add_member(:instance_health_checks, Shapes::ShapeRef.new(shape: InstanceHealthChecksRequest, location_name: "instanceHealthChecks"))
329
351
  CreateComputeNodeGroupRequest.add_member(:client_token, Shapes::ShapeRef.new(shape: SBClientToken, location_name: "clientToken", metadata: {"idempotencyToken" => true}))
330
352
  CreateComputeNodeGroupRequest.add_member(:tags, Shapes::ShapeRef.new(shape: RequestTagMap, location_name: "tags"))
331
353
  CreateComputeNodeGroupRequest.struct_class = Types::CreateComputeNodeGroupRequest
@@ -348,6 +370,13 @@ module Aws::PCS
348
370
  CustomLaunchTemplate.add_member(:version, Shapes::ShapeRef.new(shape: String, required: true, location_name: "version"))
349
371
  CustomLaunchTemplate.struct_class = Types::CustomLaunchTemplate
350
372
 
373
+ DeepHealthCheckOnStart.add_member(:profile, Shapes::ShapeRef.new(shape: DeepHealthCheckOnStartProfile, location_name: "profile"))
374
+ DeepHealthCheckOnStart.add_member(:repair_action, Shapes::ShapeRef.new(shape: DeepHealthCheckOnStartRepairAction, location_name: "repairAction"))
375
+ DeepHealthCheckOnStart.struct_class = Types::DeepHealthCheckOnStart
376
+
377
+ DeepHealthChecks.add_member(:on_start, Shapes::ShapeRef.new(shape: DeepHealthCheckOnStart, location_name: "onStart"))
378
+ DeepHealthChecks.struct_class = Types::DeepHealthChecks
379
+
351
380
  DeleteClusterRequest.add_member(:cluster_identifier, Shapes::ShapeRef.new(shape: ClusterIdentifier, required: true, location_name: "clusterIdentifier"))
352
381
  DeleteClusterRequest.add_member(:client_token, Shapes::ShapeRef.new(shape: SBClientToken, location_name: "clientToken", metadata: {"idempotencyToken" => true}))
353
382
  DeleteClusterRequest.struct_class = Types::DeleteClusterRequest
@@ -408,14 +437,38 @@ module Aws::PCS
408
437
 
409
438
  GresCustomSettings.member = Shapes::ShapeRef.new(shape: GresCustomSettingMap)
410
439
 
440
+ HealthMonitoring.add_member(:mode, Shapes::ShapeRef.new(shape: HealthMonitoringMode, location_name: "mode"))
441
+ HealthMonitoring.add_member(:repair_action, Shapes::ShapeRef.new(shape: HealthMonitoringRepairAction, location_name: "repairAction"))
442
+ HealthMonitoring.struct_class = Types::HealthMonitoring
443
+
411
444
  InstanceConfig.add_member(:instance_type, Shapes::ShapeRef.new(shape: String, location_name: "instanceType"))
412
445
  InstanceConfig.struct_class = Types::InstanceConfig
413
446
 
447
+ InstanceHealthChecks.add_member(:health_monitoring, Shapes::ShapeRef.new(shape: HealthMonitoring, location_name: "healthMonitoring"))
448
+ InstanceHealthChecks.add_member(:deep_health_checks, Shapes::ShapeRef.new(shape: DeepHealthChecks, location_name: "deepHealthChecks"))
449
+ InstanceHealthChecks.struct_class = Types::InstanceHealthChecks
450
+
451
+ InstanceHealthChecksRequest.add_member(:health_monitoring, Shapes::ShapeRef.new(shape: HealthMonitoring, location_name: "healthMonitoring"))
452
+ InstanceHealthChecksRequest.add_member(:deep_health_checks, Shapes::ShapeRef.new(shape: DeepHealthChecks, location_name: "deepHealthChecks"))
453
+ InstanceHealthChecksRequest.struct_class = Types::InstanceHealthChecksRequest
454
+
414
455
  InstanceList.member = Shapes::ShapeRef.new(shape: InstanceConfig)
415
456
 
416
457
  InternalServerException.add_member(:message, Shapes::ShapeRef.new(shape: String, required: true, location_name: "message"))
417
458
  InternalServerException.struct_class = Types::InternalServerException
418
459
 
460
+ JobSubmitPlugin.add_member(:name, Shapes::ShapeRef.new(shape: JobSubmitPluginNameString, required: true, location_name: "name"))
461
+ JobSubmitPlugin.add_member(:type, Shapes::ShapeRef.new(shape: JobSubmitPluginType, required: true, location_name: "type"))
462
+ JobSubmitPlugin.add_member(:script_source, Shapes::ShapeRef.new(shape: JobSubmitScriptSource, required: true, location_name: "scriptSource"))
463
+ JobSubmitPlugin.struct_class = Types::JobSubmitPlugin
464
+
465
+ JobSubmitPluginList.member = Shapes::ShapeRef.new(shape: JobSubmitPlugin)
466
+
467
+ JobSubmitScriptSource.add_member(:script_location, Shapes::ShapeRef.new(shape: JobSubmitScriptSourceScriptLocationString, required: true, location_name: "scriptLocation"))
468
+ JobSubmitScriptSource.add_member(:s3_version_id, Shapes::ShapeRef.new(shape: JobSubmitScriptSourceS3VersionIdString, location_name: "s3VersionId"))
469
+ JobSubmitScriptSource.add_member(:checksum, Shapes::ShapeRef.new(shape: JobSubmitScriptSourceChecksumString, location_name: "checksum"))
470
+ JobSubmitScriptSource.struct_class = Types::JobSubmitScriptSource
471
+
419
472
  JwtAuth.add_member(:jwt_key, Shapes::ShapeRef.new(shape: JwtKey, location_name: "jwtKey"))
420
473
  JwtAuth.struct_class = Types::JwtAuth
421
474
 
@@ -529,6 +582,7 @@ module Aws::PCS
529
582
  RegisterComputeNodeGroupInstanceResponse.add_member(:compute_node_group_name, Shapes::ShapeRef.new(shape: String, location_name: "computeNodeGroupName"))
530
583
  RegisterComputeNodeGroupInstanceResponse.add_member(:node_lifecycle_actions, Shapes::ShapeRef.new(shape: NodeLifecycleActions, location_name: "nodeLifecycleActions"))
531
584
  RegisterComputeNodeGroupInstanceResponse.add_member(:scheduler_version, Shapes::ShapeRef.new(shape: String, location_name: "schedulerVersion"))
585
+ RegisterComputeNodeGroupInstanceResponse.add_member(:instance_health_checks, Shapes::ShapeRef.new(shape: InstanceHealthChecks, location_name: "instanceHealthChecks"))
532
586
  RegisterComputeNodeGroupInstanceResponse.struct_class = Types::RegisterComputeNodeGroupInstanceResponse
533
587
 
534
588
  RequestTagMap.key = Shapes::ShapeRef.new(shape: TagKey)
@@ -638,6 +692,7 @@ module Aws::PCS
638
692
  UpdateClusterSlurmConfigurationRequest.add_member(:slurm_custom_settings, Shapes::ShapeRef.new(shape: SlurmCustomSettings, location_name: "slurmCustomSettings"))
639
693
  UpdateClusterSlurmConfigurationRequest.add_member(:slurmdbd_custom_settings, Shapes::ShapeRef.new(shape: SlurmdbdCustomSettings, location_name: "slurmdbdCustomSettings"))
640
694
  UpdateClusterSlurmConfigurationRequest.add_member(:cgroup_custom_settings, Shapes::ShapeRef.new(shape: CgroupCustomSettings, location_name: "cgroupCustomSettings"))
695
+ UpdateClusterSlurmConfigurationRequest.add_member(:job_submit_plugins, Shapes::ShapeRef.new(shape: JobSubmitPluginList, location_name: "jobSubmitPlugins"))
641
696
  UpdateClusterSlurmConfigurationRequest.add_member(:accounting, Shapes::ShapeRef.new(shape: UpdateAccountingRequest, location_name: "accounting"))
642
697
  UpdateClusterSlurmConfigurationRequest.add_member(:slurm_rest, Shapes::ShapeRef.new(shape: UpdateSlurmRestRequest, location_name: "slurmRest"))
643
698
  UpdateClusterSlurmConfigurationRequest.struct_class = Types::UpdateClusterSlurmConfigurationRequest
@@ -656,6 +711,7 @@ module Aws::PCS
656
711
  UpdateComputeNodeGroupRequest.add_member(:iam_instance_profile_arn, Shapes::ShapeRef.new(shape: InstanceProfileArn, location_name: "iamInstanceProfileArn"))
657
712
  UpdateComputeNodeGroupRequest.add_member(:slurm_configuration, Shapes::ShapeRef.new(shape: UpdateComputeNodeGroupSlurmConfigurationRequest, location_name: "slurmConfiguration"))
658
713
  UpdateComputeNodeGroupRequest.add_member(:node_lifecycle_actions, Shapes::ShapeRef.new(shape: UpdateNodeLifecycleActionsRequest, location_name: "nodeLifecycleActions"))
714
+ UpdateComputeNodeGroupRequest.add_member(:instance_health_checks, Shapes::ShapeRef.new(shape: UpdateInstanceHealthChecksRequest, location_name: "instanceHealthChecks"))
659
715
  UpdateComputeNodeGroupRequest.add_member(:client_token, Shapes::ShapeRef.new(shape: SBClientToken, location_name: "clientToken", metadata: {"idempotencyToken" => true}))
660
716
  UpdateComputeNodeGroupRequest.struct_class = Types::UpdateComputeNodeGroupRequest
661
717
 
@@ -667,6 +723,10 @@ module Aws::PCS
667
723
  UpdateComputeNodeGroupSlurmConfigurationRequest.add_member(:gres_custom_settings, Shapes::ShapeRef.new(shape: GresCustomSettings, location_name: "gresCustomSettings"))
668
724
  UpdateComputeNodeGroupSlurmConfigurationRequest.struct_class = Types::UpdateComputeNodeGroupSlurmConfigurationRequest
669
725
 
726
+ UpdateInstanceHealthChecksRequest.add_member(:health_monitoring, Shapes::ShapeRef.new(shape: HealthMonitoring, location_name: "healthMonitoring"))
727
+ UpdateInstanceHealthChecksRequest.add_member(:deep_health_checks, Shapes::ShapeRef.new(shape: DeepHealthChecks, location_name: "deepHealthChecks"))
728
+ UpdateInstanceHealthChecksRequest.struct_class = Types::UpdateInstanceHealthChecksRequest
729
+
670
730
  UpdateNodeLifecycleActionsRequest.add_member(:stages, Shapes::ShapeRef.new(shape: NodeLifecycleStages, required: true, location_name: "stages"))
671
731
  UpdateNodeLifecycleActionsRequest.add_member(:script_caching_policy, Shapes::ShapeRef.new(shape: ScriptCachingPolicy, location_name: "scriptCachingPolicy"))
672
732
  UpdateNodeLifecycleActionsRequest.struct_class = Types::UpdateNodeLifecycleActionsRequest
@@ -254,6 +254,14 @@ module Aws::PCS
254
254
  # Cgroup settings.
255
255
  # @return [Array<Types::CgroupCustomSetting>]
256
256
  #
257
+ # @!attribute [rw] job_submit_plugins
258
+ # The job submit plugins to run on the cluster controller. Use job
259
+ # submit plugins to execute custom scripts during job submission and
260
+ # job modification to validate or modify the job's configuration.
261
+ # Custom scripts can accept or reject jobs based on their
262
+ # configuration or modify the configuration before job execution.
263
+ # @return [Array<Types::JobSubmitPlugin>]
264
+ #
257
265
  # @!attribute [rw] auth_key
258
266
  # The shared Slurm key for authentication, also known as the **cluster
259
267
  # secret**.
@@ -279,6 +287,7 @@ module Aws::PCS
279
287
  :slurm_custom_settings,
280
288
  :slurmdbd_custom_settings,
281
289
  :cgroup_custom_settings,
290
+ :job_submit_plugins,
282
291
  :auth_key,
283
292
  :jwt_auth,
284
293
  :accounting,
@@ -310,6 +319,14 @@ module Aws::PCS
310
319
  # Cgroup settings.
311
320
  # @return [Array<Types::CgroupCustomSetting>]
312
321
  #
322
+ # @!attribute [rw] job_submit_plugins
323
+ # The job submit plugins to run on the cluster controller. Use job
324
+ # submit plugins to execute custom scripts during job submission and
325
+ # job modification to validate or modify the job's configuration.
326
+ # Custom scripts can accept or reject jobs based on their
327
+ # configuration or modify the configuration before job execution.
328
+ # @return [Array<Types::JobSubmitPlugin>]
329
+ #
313
330
  # @!attribute [rw] accounting
314
331
  # The accounting configuration includes configurable settings for
315
332
  # Slurm accounting.
@@ -326,6 +343,7 @@ module Aws::PCS
326
343
  :slurm_custom_settings,
327
344
  :slurmdbd_custom_settings,
328
345
  :cgroup_custom_settings,
346
+ :job_submit_plugins,
329
347
  :accounting,
330
348
  :slurm_rest)
331
349
  SENSITIVE = []
@@ -548,6 +566,12 @@ module Aws::PCS
548
566
  # bootstrapping or becomes ready to accept jobs.
549
567
  # @return [Types::NodeLifecycleActions]
550
568
  #
569
+ # @!attribute [rw] instance_health_checks
570
+ # The instance health check settings of the compute node group.
571
+ # Instance health checks detect faults on instances in the compute
572
+ # node group and respond with the repair action that you choose.
573
+ # @return [Types::InstanceHealthChecks]
574
+ #
551
575
  # @!attribute [rw] error_info
552
576
  # The list of errors that occurred during compute node group
553
577
  # provisioning.
@@ -576,6 +600,7 @@ module Aws::PCS
576
600
  :spot_options,
577
601
  :slurm_configuration,
578
602
  :node_lifecycle_actions,
603
+ :instance_health_checks,
579
604
  :error_info)
580
605
  SENSITIVE = []
581
606
  include Aws::Structure
@@ -967,6 +992,15 @@ module Aws::PCS
967
992
  # bootstrapping or becomes ready to accept jobs.
968
993
  # @return [Types::NodeLifecycleActionsRequest]
969
994
  #
995
+ # @!attribute [rw] instance_health_checks
996
+ # The instance health check settings for the compute node group.
997
+ # Instance health checks detect faults on instances in the compute
998
+ # node group and respond with the repair action that you choose. You
999
+ # can configure health monitoring, deep health checks, or both. You
1000
+ # can't specify an empty object (`{}`). Instance health checks
1001
+ # require Slurm version 26.05 or later.
1002
+ # @return [Types::InstanceHealthChecksRequest]
1003
+ #
970
1004
  # @!attribute [rw] client_token
971
1005
  # A unique, case-sensitive identifier that you provide to ensure the
972
1006
  # idempotency of the request. Idempotency ensures that an API request
@@ -1003,6 +1037,7 @@ module Aws::PCS
1003
1037
  :spot_options,
1004
1038
  :slurm_configuration,
1005
1039
  :node_lifecycle_actions,
1040
+ :instance_health_checks,
1006
1041
  :client_token,
1007
1042
  :tags)
1008
1043
  SENSITIVE = []
@@ -1105,6 +1140,83 @@ module Aws::PCS
1105
1140
  include Aws::Structure
1106
1141
  end
1107
1142
 
1143
+ # The deep health check that runs each time an instance starts,
1144
+ # including after a reboot, before the compute node accepts jobs.
1145
+ #
1146
+ # @!attribute [rw] profile
1147
+ # The deep health check profile to run. Each profile other than `NONE`
1148
+ # runs NVIDIA Data Center GPU Manager (DCGM) diagnostics. More
1149
+ # thorough profiles run more tests and add more time to instance
1150
+ # launch. A profile other than `NONE` requires instance types with
1151
+ # NVIDIA GPUs. This value is required when you specify `onStart`.
1152
+ # There is no default value. Valid values:
1153
+ #
1154
+ # * `NONE` – Turns off the deep health check. You can't specify
1155
+ # `repairAction` with this profile.
1156
+ #
1157
+ # * `QUICK` – Runs basic GPU tests (DCGM diagnostics level 1). This
1158
+ # profile adds the least time to instance launch.
1159
+ #
1160
+ # * `MEDIUM` – Runs extended GPU tests (DCGM diagnostics level 2).
1161
+ #
1162
+ # * `LONG` – Runs GPU stress and diagnostic tests (DCGM diagnostics
1163
+ # level 3). This profile adds the most time to instance launch.
1164
+ # @return [String]
1165
+ #
1166
+ # @!attribute [rw] repair_action
1167
+ # The action that PCS takes when an instance fails the deep health
1168
+ # check. You can't specify this value when `profile` is `NONE`. If
1169
+ # you don't specify this value and `profile` isn't `NONE`, PCS uses
1170
+ # `AUTO`. `GetComputeNodeGroup` doesn't return a `repairAction` that
1171
+ # you didn't specify. Valid values:
1172
+ #
1173
+ # * `AUTO` – PCS chooses the repair action based on the fault.
1174
+ #
1175
+ # * `REPLACE` – Replaces the instance before it accepts jobs.
1176
+ #
1177
+ # * `ISOLATE` – Stops new jobs on the instance and keeps the instance
1178
+ # running until you investigate.
1179
+ #
1180
+ # * `NONE` – Records the fault in Slurm and leaves the instance
1181
+ # available for jobs.
1182
+ # @return [String]
1183
+ #
1184
+ # @see http://docs.aws.amazon.com/goto/WebAPI/pcs-2023-02-10/DeepHealthCheckOnStart AWS API Documentation
1185
+ #
1186
+ class DeepHealthCheckOnStart < Struct.new(
1187
+ :profile,
1188
+ :repair_action)
1189
+ SENSITIVE = []
1190
+ include Aws::Structure
1191
+ end
1192
+
1193
+ # The deep health check settings of a compute node group. Deep health
1194
+ # checks test the graphics processing units (GPUs) of an instance each
1195
+ # time it starts, including after a reboot, before it accepts jobs. Deep
1196
+ # health checks with a profile other than `NONE` require instance types
1197
+ # with NVIDIA GPUs.
1198
+ #
1199
+ # @!attribute [rw] on_start
1200
+ # The deep health check to run each time an instance starts, including
1201
+ # after a reboot, before the compute node accepts jobs. If the check
1202
+ # fails after a reboot, PCS takes the same `repairAction` as at
1203
+ # launch. For example, with `REPLACE`, PCS replaces the instance
1204
+ # instead of returning it to service. If the compute node group has
1205
+ # scripts in the `nodeBootstrapped` stage of `nodeLifecycleActions`,
1206
+ # the deep health check runs after those scripts. If a script fails
1207
+ # and its `onError` setting terminates the compute node, the deep
1208
+ # health check doesn't run. This value is required when you specify
1209
+ # `deepHealthChecks`.
1210
+ # @return [Types::DeepHealthCheckOnStart]
1211
+ #
1212
+ # @see http://docs.aws.amazon.com/goto/WebAPI/pcs-2023-02-10/DeepHealthChecks AWS API Documentation
1213
+ #
1214
+ class DeepHealthChecks < Struct.new(
1215
+ :on_start)
1216
+ SENSITIVE = []
1217
+ include Aws::Structure
1218
+ end
1219
+
1108
1220
  # @!attribute [rw] cluster_identifier
1109
1221
  # The name or ID of the cluster to delete.
1110
1222
  # @return [String]
@@ -1351,6 +1463,48 @@ module Aws::PCS
1351
1463
  include Aws::Structure
1352
1464
  end
1353
1465
 
1466
+ # The health monitoring settings of a compute node group. Health
1467
+ # monitoring continuously checks running instances for hardware and
1468
+ # software faults.
1469
+ #
1470
+ # @!attribute [rw] mode
1471
+ # The health monitoring mode. This value is required when you specify
1472
+ # `healthMonitoring`. There is no default value. Valid values:
1473
+ #
1474
+ # * `NONE` – Turns off health monitoring. You can't specify
1475
+ # `repairAction` with this mode.
1476
+ #
1477
+ # * `STANDARD` – Continuously checks running instances for hardware
1478
+ # and software faults.
1479
+ # @return [String]
1480
+ #
1481
+ # @!attribute [rw] repair_action
1482
+ # The action that PCS takes when health monitoring finds a fault on an
1483
+ # instance. You can't specify this value when `mode` is `NONE`. If
1484
+ # you don't specify this value and `mode` isn't `NONE`, PCS uses
1485
+ # `AUTO`. `GetComputeNodeGroup` doesn't return a `repairAction` that
1486
+ # you didn't specify. Valid values:
1487
+ #
1488
+ # * `AUTO` – PCS chooses the repair action based on the fault.
1489
+ #
1490
+ # * `REPLACE` – Replaces the instance after running jobs finish.
1491
+ #
1492
+ # * `ISOLATE` – Stops new jobs on the instance and keeps the instance
1493
+ # running until you investigate.
1494
+ #
1495
+ # * `NONE` – Records the fault in Slurm and leaves the instance
1496
+ # available for jobs.
1497
+ # @return [String]
1498
+ #
1499
+ # @see http://docs.aws.amazon.com/goto/WebAPI/pcs-2023-02-10/HealthMonitoring AWS API Documentation
1500
+ #
1501
+ class HealthMonitoring < Struct.new(
1502
+ :mode,
1503
+ :repair_action)
1504
+ SENSITIVE = []
1505
+ include Aws::Structure
1506
+ end
1507
+
1354
1508
  # An EC2 instance configuration PCS uses to launch compute nodes.
1355
1509
  #
1356
1510
  # @!attribute [rw] instance_type
@@ -1368,6 +1522,57 @@ module Aws::PCS
1368
1522
  include Aws::Structure
1369
1523
  end
1370
1524
 
1525
+ # The instance health check settings of a compute node group. Instance
1526
+ # health checks detect faults on instances in the compute node group and
1527
+ # respond with the repair action that you choose.
1528
+ #
1529
+ # @!attribute [rw] health_monitoring
1530
+ # The health monitoring settings. Health monitoring continuously
1531
+ # checks running instances for hardware and software faults.
1532
+ # @return [Types::HealthMonitoring]
1533
+ #
1534
+ # @!attribute [rw] deep_health_checks
1535
+ # The deep health check settings. Deep health checks test the graphics
1536
+ # processing units (GPUs) of an instance each time it starts,
1537
+ # including after a reboot, before it accepts jobs.
1538
+ # @return [Types::DeepHealthChecks]
1539
+ #
1540
+ # @see http://docs.aws.amazon.com/goto/WebAPI/pcs-2023-02-10/InstanceHealthChecks AWS API Documentation
1541
+ #
1542
+ class InstanceHealthChecks < Struct.new(
1543
+ :health_monitoring,
1544
+ :deep_health_checks)
1545
+ SENSITIVE = []
1546
+ include Aws::Structure
1547
+ end
1548
+
1549
+ # The instance health check settings to configure on a compute node
1550
+ # group when you create it. Specify `healthMonitoring`,
1551
+ # `deepHealthChecks`, or both. You can't specify an empty object
1552
+ # (`{}`).
1553
+ #
1554
+ # @!attribute [rw] health_monitoring
1555
+ # The health monitoring settings. Health monitoring continuously
1556
+ # checks running instances for hardware and software faults. If you
1557
+ # specify this object, it can't be empty.
1558
+ # @return [Types::HealthMonitoring]
1559
+ #
1560
+ # @!attribute [rw] deep_health_checks
1561
+ # The deep health check settings. Deep health checks test the graphics
1562
+ # processing units (GPUs) of an instance each time it starts,
1563
+ # including after a reboot, before it accepts jobs. If you specify
1564
+ # this object, it can't be empty.
1565
+ # @return [Types::DeepHealthChecks]
1566
+ #
1567
+ # @see http://docs.aws.amazon.com/goto/WebAPI/pcs-2023-02-10/InstanceHealthChecksRequest AWS API Documentation
1568
+ #
1569
+ class InstanceHealthChecksRequest < Struct.new(
1570
+ :health_monitoring,
1571
+ :deep_health_checks)
1572
+ SENSITIVE = []
1573
+ include Aws::Structure
1574
+ end
1575
+
1371
1576
  # PCS can't process your request right now. Try again later.
1372
1577
  #
1373
1578
  # @!attribute [rw] message
@@ -1381,6 +1586,67 @@ module Aws::PCS
1381
1586
  include Aws::Structure
1382
1587
  end
1383
1588
 
1589
+ # A job submit plugin configuration, including its name, runtime
1590
+ # language, and script source.
1591
+ #
1592
+ # @!attribute [rw] name
1593
+ # A unique name for the plugin. The name can be up to 64 characters
1594
+ # long. Valid characters are letters, numbers, spaces, underscores
1595
+ # (`_`), and hyphens (`-`). The first character must be a letter or a
1596
+ # number.
1597
+ # @return [String]
1598
+ #
1599
+ # @!attribute [rw] type
1600
+ # The runtime language used to execute the custom script. Valid
1601
+ # values:
1602
+ #
1603
+ # * `STARLARK` – The script uses Starlark syntax.
1604
+ #
1605
+ # ^
1606
+ # @return [String]
1607
+ #
1608
+ # @!attribute [rw] script_source
1609
+ # The source location and integrity information for the script.
1610
+ # @return [Types::JobSubmitScriptSource]
1611
+ #
1612
+ # @see http://docs.aws.amazon.com/goto/WebAPI/pcs-2023-02-10/JobSubmitPlugin AWS API Documentation
1613
+ #
1614
+ class JobSubmitPlugin < Struct.new(
1615
+ :name,
1616
+ :type,
1617
+ :script_source)
1618
+ SENSITIVE = []
1619
+ include Aws::Structure
1620
+ end
1621
+
1622
+ # The source location and integrity information for the script.
1623
+ #
1624
+ # @!attribute [rw] script_location
1625
+ # The location of the script. Specify an Amazon S3 URI in the format
1626
+ # `s3://bucket-name/key`.
1627
+ # @return [String]
1628
+ #
1629
+ # @!attribute [rw] s3_version_id
1630
+ # The Amazon S3 version ID of the script. Use this value to pin the
1631
+ # script to a specific version in a versioned Amazon S3 bucket.
1632
+ # @return [String]
1633
+ #
1634
+ # @!attribute [rw] checksum
1635
+ # The SHA-256 checksum of the script content, as a 64-character
1636
+ # hexadecimal string. This value is optional. When specified, PCS uses
1637
+ # this value to verify the integrity of the downloaded script.
1638
+ # @return [String]
1639
+ #
1640
+ # @see http://docs.aws.amazon.com/goto/WebAPI/pcs-2023-02-10/JobSubmitScriptSource AWS API Documentation
1641
+ #
1642
+ class JobSubmitScriptSource < Struct.new(
1643
+ :script_location,
1644
+ :s3_version_id,
1645
+ :checksum)
1646
+ SENSITIVE = []
1647
+ include Aws::Structure
1648
+ end
1649
+
1384
1650
  # The JWT authentication configuration for Slurm REST API access.
1385
1651
  #
1386
1652
  # @!attribute [rw] jwt_key
@@ -2045,6 +2311,13 @@ module Aws::PCS
2045
2311
  # uses this value to run the matching scheduler version.
2046
2312
  # @return [String]
2047
2313
  #
2314
+ # @!attribute [rw] instance_health_checks
2315
+ # The instance health check settings configured for the compute node
2316
+ # group that the compute node registered into. The compute node uses
2317
+ # these settings to configure health monitoring and deep health
2318
+ # checks.
2319
+ # @return [Types::InstanceHealthChecks]
2320
+ #
2048
2321
  # @see http://docs.aws.amazon.com/goto/WebAPI/pcs-2023-02-10/RegisterComputeNodeGroupInstanceResponse AWS API Documentation
2049
2322
  #
2050
2323
  class RegisterComputeNodeGroupInstanceResponse < Struct.new(
@@ -2055,7 +2328,8 @@ module Aws::PCS
2055
2328
  :compute_node_group_id,
2056
2329
  :compute_node_group_name,
2057
2330
  :node_lifecycle_actions,
2058
- :scheduler_version)
2331
+ :scheduler_version,
2332
+ :instance_health_checks)
2059
2333
  SENSITIVE = [:shared_secret]
2060
2334
  include Aws::Structure
2061
2335
  end
@@ -2590,6 +2864,14 @@ module Aws::PCS
2590
2864
  # Cgroup settings.
2591
2865
  # @return [Array<Types::CgroupCustomSetting>]
2592
2866
  #
2867
+ # @!attribute [rw] job_submit_plugins
2868
+ # The job submit plugins to run on the cluster controller. Use job
2869
+ # submit plugins to execute custom scripts during job submission and
2870
+ # job modification to validate or modify the job's configuration.
2871
+ # Custom scripts can accept or reject jobs based on their
2872
+ # configuration or modify the configuration before job execution.
2873
+ # @return [Array<Types::JobSubmitPlugin>]
2874
+ #
2593
2875
  # @!attribute [rw] accounting
2594
2876
  # The accounting configuration includes configurable settings for
2595
2877
  # Slurm accounting.
@@ -2606,6 +2888,7 @@ module Aws::PCS
2606
2888
  :slurm_custom_settings,
2607
2889
  :slurmdbd_custom_settings,
2608
2890
  :cgroup_custom_settings,
2891
+ :job_submit_plugins,
2609
2892
  :accounting,
2610
2893
  :slurm_rest)
2611
2894
  SENSITIVE = []
@@ -2754,6 +3037,25 @@ module Aws::PCS
2754
3037
  # bootstrapping or becomes ready to accept jobs.
2755
3038
  # @return [Types::UpdateNodeLifecycleActionsRequest]
2756
3039
  #
3040
+ # @!attribute [rw] instance_health_checks
3041
+ # The instance health check settings for the compute node group.
3042
+ # Instance health checks detect faults on instances in the compute
3043
+ # node group and respond with the repair action that you choose. You
3044
+ # can configure health monitoring, deep health checks, or both. You
3045
+ # can't specify an empty object (`{}`). Instance health checks
3046
+ # require Slurm version 26.05 or later.
3047
+ #
3048
+ # The settings that you specify replace all of the current instance
3049
+ # health check settings of the compute node group. Include every
3050
+ # setting that you want to keep. If you omit this member, the current
3051
+ # settings don't change. If you omit `healthMonitoring` or
3052
+ # `deepHealthChecks`, that check is turned off. You can also turn off
3053
+ # a check: set `healthMonitoring.mode` or
3054
+ # `deepHealthChecks.onStart.profile` to `NONE` and omit its
3055
+ # `repairAction`. When you change these settings, PCS drains and
3056
+ # replaces the instances in the compute node group.
3057
+ # @return [Types::UpdateInstanceHealthChecksRequest]
3058
+ #
2757
3059
  # @!attribute [rw] client_token
2758
3060
  # A unique, case-sensitive identifier that you provide to ensure the
2759
3061
  # idempotency of the request. Idempotency ensures that an API request
@@ -2784,6 +3086,7 @@ module Aws::PCS
2784
3086
  :iam_instance_profile_arn,
2785
3087
  :slurm_configuration,
2786
3088
  :node_lifecycle_actions,
3089
+ :instance_health_checks,
2787
3090
  :client_token)
2788
3091
  SENSITIVE = []
2789
3092
  include Aws::Structure
@@ -2834,6 +3137,36 @@ module Aws::PCS
2834
3137
  include Aws::Structure
2835
3138
  end
2836
3139
 
3140
+ # The instance health check settings to configure on a compute node
3141
+ # group when you update it. These settings replace all of the current
3142
+ # instance health check settings of the compute node group. Specify
3143
+ # `healthMonitoring`, `deepHealthChecks`, or both. You can't specify an
3144
+ # empty object (`{}`).
3145
+ #
3146
+ # @!attribute [rw] health_monitoring
3147
+ # The health monitoring settings. Health monitoring continuously
3148
+ # checks running instances for hardware and software faults. If you
3149
+ # specify this object, it can't be empty. If you omit this object,
3150
+ # health monitoring is turned off.
3151
+ # @return [Types::HealthMonitoring]
3152
+ #
3153
+ # @!attribute [rw] deep_health_checks
3154
+ # The deep health check settings. Deep health checks test the graphics
3155
+ # processing units (GPUs) of an instance each time it starts,
3156
+ # including after a reboot, before it accepts jobs. If you specify
3157
+ # this object, it can't be empty. If you omit this object, deep
3158
+ # health checks are turned off.
3159
+ # @return [Types::DeepHealthChecks]
3160
+ #
3161
+ # @see http://docs.aws.amazon.com/goto/WebAPI/pcs-2023-02-10/UpdateInstanceHealthChecksRequest AWS API Documentation
3162
+ #
3163
+ class UpdateInstanceHealthChecksRequest < Struct.new(
3164
+ :health_monitoring,
3165
+ :deep_health_checks)
3166
+ SENSITIVE = []
3167
+ include Aws::Structure
3168
+ end
3169
+
2837
3170
  # The lifecycle actions to configure on a compute node group when you
2838
3171
  # update it. Lifecycle actions define scripts that PCS runs on compute
2839
3172
  # nodes at specific stages of their lifecycle.
data/lib/aws-sdk-pcs.rb CHANGED
@@ -55,7 +55,7 @@ module Aws::PCS
55
55
  autoload :EndpointProvider, 'aws-sdk-pcs/endpoint_provider'
56
56
  autoload :Endpoints, 'aws-sdk-pcs/endpoints'
57
57
 
58
- GEM_VERSION = '1.67.0'
58
+ GEM_VERSION = '1.69.0'
59
59
 
60
60
  end
61
61
 
data/sig/client.rbs CHANGED
@@ -117,6 +117,9 @@ module Aws
117
117
  parameter_value: ::String
118
118
  }
119
119
  ]?,
120
+ job_submit_plugins: Array[
121
+ Params::job_submit_plugin
122
+ ]?,
120
123
  accounting: {
121
124
  default_purge_time_in_days: ::Integer?,
122
125
  mode: ("STANDARD" | "NONE")
@@ -177,6 +180,18 @@ module Aws
177
180
  stages: Params::node_lifecycle_stages,
178
181
  script_caching_policy: ("CACHE_ONCE" | "REFRESH_ON_REBOOT")?
179
182
  },
183
+ ?instance_health_checks: {
184
+ health_monitoring: {
185
+ mode: ("NONE" | "STANDARD")?,
186
+ repair_action: ("AUTO" | "REPLACE" | "ISOLATE" | "NONE")?
187
+ }?,
188
+ deep_health_checks: {
189
+ on_start: {
190
+ profile: ("NONE" | "QUICK" | "MEDIUM" | "LONG")?,
191
+ repair_action: ("AUTO" | "REPLACE" | "ISOLATE" | "NONE")?
192
+ }?
193
+ }?
194
+ },
180
195
  ?client_token: ::String,
181
196
  ?tags: Hash[::String, ::String]
182
197
  ) -> _CreateComputeNodeGroupResponseSuccess
@@ -330,6 +345,7 @@ module Aws
330
345
  def compute_node_group_name: () -> ::String
331
346
  def node_lifecycle_actions: () -> Types::NodeLifecycleActions
332
347
  def scheduler_version: () -> ::String
348
+ def instance_health_checks: () -> Types::InstanceHealthChecks
333
349
  end
334
350
  # https://docs.aws.amazon.com/sdk-for-ruby/v3/api/Aws/PCS/Client.html#register_compute_node_group_instance-instance_method
335
351
  def register_compute_node_group_instance: (
@@ -386,6 +402,9 @@ module Aws
386
402
  parameter_value: ::String
387
403
  }
388
404
  ]?,
405
+ job_submit_plugins: Array[
406
+ Params::job_submit_plugin
407
+ ]?,
389
408
  accounting: {
390
409
  default_purge_time_in_days: ::Integer?,
391
410
  mode: ("STANDARD" | "NONE")?
@@ -442,6 +461,18 @@ module Aws
442
461
  stages: Params::node_lifecycle_stages,
443
462
  script_caching_policy: ("CACHE_ONCE" | "REFRESH_ON_REBOOT")?
444
463
  },
464
+ ?instance_health_checks: {
465
+ health_monitoring: {
466
+ mode: ("NONE" | "STANDARD")?,
467
+ repair_action: ("AUTO" | "REPLACE" | "ISOLATE" | "NONE")?
468
+ }?,
469
+ deep_health_checks: {
470
+ on_start: {
471
+ profile: ("NONE" | "QUICK" | "MEDIUM" | "LONG")?,
472
+ repair_action: ("AUTO" | "REPLACE" | "ISOLATE" | "NONE")?
473
+ }?
474
+ }?
475
+ },
445
476
  ?client_token: ::String
446
477
  ) -> _UpdateComputeNodeGroupResponseSuccess
447
478
  | (Hash[Symbol, untyped] params, ?Hash[Symbol, untyped] options) -> _UpdateComputeNodeGroupResponseSuccess
data/sig/params.rbs CHANGED
@@ -8,6 +8,16 @@
8
8
  module Aws
9
9
  module PCS
10
10
  module Params
11
+ type job_submit_plugin = {
12
+ name: ::String,
13
+ type: ("STARLARK"),
14
+ script_source: {
15
+ script_location: ::String,
16
+ s3_version_id: ::String?,
17
+ checksum: ::String?
18
+ }
19
+ }
20
+
11
21
  type node_lifecycle_stages = {
12
22
  node_bootstrapped: Array[
13
23
  Params::node_lifecycle_script
data/sig/types.rbs CHANGED
@@ -52,6 +52,7 @@ module Aws::PCS
52
52
  attr_accessor slurm_custom_settings: ::Array[Types::SlurmCustomSetting]
53
53
  attr_accessor slurmdbd_custom_settings: ::Array[Types::SlurmdbdCustomSetting]
54
54
  attr_accessor cgroup_custom_settings: ::Array[Types::CgroupCustomSetting]
55
+ attr_accessor job_submit_plugins: ::Array[Types::JobSubmitPlugin]
55
56
  attr_accessor auth_key: Types::SlurmAuthKey
56
57
  attr_accessor jwt_auth: Types::JwtAuth
57
58
  attr_accessor accounting: Types::Accounting
@@ -64,6 +65,7 @@ module Aws::PCS
64
65
  attr_accessor slurm_custom_settings: ::Array[Types::SlurmCustomSetting]
65
66
  attr_accessor slurmdbd_custom_settings: ::Array[Types::SlurmdbdCustomSetting]
66
67
  attr_accessor cgroup_custom_settings: ::Array[Types::CgroupCustomSetting]
68
+ attr_accessor job_submit_plugins: ::Array[Types::JobSubmitPlugin]
67
69
  attr_accessor accounting: Types::AccountingRequest
68
70
  attr_accessor slurm_rest: Types::SlurmRestRequest
69
71
  SENSITIVE: []
@@ -100,6 +102,7 @@ module Aws::PCS
100
102
  attr_accessor spot_options: Types::SpotOptions
101
103
  attr_accessor slurm_configuration: Types::ComputeNodeGroupSlurmConfiguration
102
104
  attr_accessor node_lifecycle_actions: Types::NodeLifecycleActions
105
+ attr_accessor instance_health_checks: Types::InstanceHealthChecks
103
106
  attr_accessor error_info: ::Array[Types::ErrorInfo]
104
107
  SENSITIVE: []
105
108
  end
@@ -173,6 +176,7 @@ module Aws::PCS
173
176
  attr_accessor spot_options: Types::SpotOptions
174
177
  attr_accessor slurm_configuration: Types::ComputeNodeGroupSlurmConfigurationRequest
175
178
  attr_accessor node_lifecycle_actions: Types::NodeLifecycleActionsRequest
179
+ attr_accessor instance_health_checks: Types::InstanceHealthChecksRequest
176
180
  attr_accessor client_token: ::String
177
181
  attr_accessor tags: ::Hash[::String, ::String]
178
182
  SENSITIVE: []
@@ -204,6 +208,17 @@ module Aws::PCS
204
208
  SENSITIVE: []
205
209
  end
206
210
 
211
+ class DeepHealthCheckOnStart
212
+ attr_accessor profile: ("NONE" | "QUICK" | "MEDIUM" | "LONG")
213
+ attr_accessor repair_action: ("AUTO" | "REPLACE" | "ISOLATE" | "NONE")
214
+ SENSITIVE: []
215
+ end
216
+
217
+ class DeepHealthChecks
218
+ attr_accessor on_start: Types::DeepHealthCheckOnStart
219
+ SENSITIVE: []
220
+ end
221
+
207
222
  class DeleteClusterRequest
208
223
  attr_accessor cluster_identifier: ::String
209
224
  attr_accessor client_token: ::String
@@ -280,16 +295,48 @@ module Aws::PCS
280
295
  SENSITIVE: []
281
296
  end
282
297
 
298
+ class HealthMonitoring
299
+ attr_accessor mode: ("NONE" | "STANDARD")
300
+ attr_accessor repair_action: ("AUTO" | "REPLACE" | "ISOLATE" | "NONE")
301
+ SENSITIVE: []
302
+ end
303
+
283
304
  class InstanceConfig
284
305
  attr_accessor instance_type: ::String
285
306
  SENSITIVE: []
286
307
  end
287
308
 
309
+ class InstanceHealthChecks
310
+ attr_accessor health_monitoring: Types::HealthMonitoring
311
+ attr_accessor deep_health_checks: Types::DeepHealthChecks
312
+ SENSITIVE: []
313
+ end
314
+
315
+ class InstanceHealthChecksRequest
316
+ attr_accessor health_monitoring: Types::HealthMonitoring
317
+ attr_accessor deep_health_checks: Types::DeepHealthChecks
318
+ SENSITIVE: []
319
+ end
320
+
288
321
  class InternalServerException
289
322
  attr_accessor message: ::String
290
323
  SENSITIVE: []
291
324
  end
292
325
 
326
+ class JobSubmitPlugin
327
+ attr_accessor name: ::String
328
+ attr_accessor type: ("STARLARK")
329
+ attr_accessor script_source: Types::JobSubmitScriptSource
330
+ SENSITIVE: []
331
+ end
332
+
333
+ class JobSubmitScriptSource
334
+ attr_accessor script_location: ::String
335
+ attr_accessor s3_version_id: ::String
336
+ attr_accessor checksum: ::String
337
+ SENSITIVE: []
338
+ end
339
+
293
340
  class JwtAuth
294
341
  attr_accessor jwt_key: Types::JwtKey
295
342
  SENSITIVE: []
@@ -440,6 +487,7 @@ module Aws::PCS
440
487
  attr_accessor compute_node_group_name: ::String
441
488
  attr_accessor node_lifecycle_actions: Types::NodeLifecycleActions
442
489
  attr_accessor scheduler_version: ::String
490
+ attr_accessor instance_health_checks: Types::InstanceHealthChecks
443
491
  SENSITIVE: [:shared_secret]
444
492
  end
445
493
 
@@ -571,6 +619,7 @@ module Aws::PCS
571
619
  attr_accessor slurm_custom_settings: ::Array[Types::SlurmCustomSetting]
572
620
  attr_accessor slurmdbd_custom_settings: ::Array[Types::SlurmdbdCustomSetting]
573
621
  attr_accessor cgroup_custom_settings: ::Array[Types::CgroupCustomSetting]
622
+ attr_accessor job_submit_plugins: ::Array[Types::JobSubmitPlugin]
574
623
  attr_accessor accounting: Types::UpdateAccountingRequest
575
624
  attr_accessor slurm_rest: Types::UpdateSlurmRestRequest
576
625
  SENSITIVE: []
@@ -591,6 +640,7 @@ module Aws::PCS
591
640
  attr_accessor iam_instance_profile_arn: ::String
592
641
  attr_accessor slurm_configuration: Types::UpdateComputeNodeGroupSlurmConfigurationRequest
593
642
  attr_accessor node_lifecycle_actions: Types::UpdateNodeLifecycleActionsRequest
643
+ attr_accessor instance_health_checks: Types::UpdateInstanceHealthChecksRequest
594
644
  attr_accessor client_token: ::String
595
645
  SENSITIVE: []
596
646
  end
@@ -607,6 +657,12 @@ module Aws::PCS
607
657
  SENSITIVE: []
608
658
  end
609
659
 
660
+ class UpdateInstanceHealthChecksRequest
661
+ attr_accessor health_monitoring: Types::HealthMonitoring
662
+ attr_accessor deep_health_checks: Types::DeepHealthChecks
663
+ SENSITIVE: []
664
+ end
665
+
610
666
  class UpdateNodeLifecycleActionsRequest
611
667
  attr_accessor stages: Types::NodeLifecycleStages
612
668
  attr_accessor script_caching_policy: ("CACHE_ONCE" | "REFRESH_ON_REBOOT")
metadata CHANGED
@@ -1,7 +1,7 @@
1
1
  --- !ruby/object:Gem::Specification
2
2
  name: aws-sdk-pcs
3
3
  version: !ruby/object:Gem::Version
4
- version: 1.67.0
4
+ version: 1.69.0
5
5
  platform: ruby
6
6
  authors:
7
7
  - Amazon Web Services