xpk 0.17.0__tar.gz → 0.17.2__tar.gz

{xpk-0.17.0/src/xpk.egg-info → xpk-0.17.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: xpk
-Version: 0.17.0
+Version: 0.17.2
 Summary: xpk helps Cloud developers to orchestrate training jobs on accelerators on GKE.
 Author-email: XPK team <xpk-code-reviewers@google.com>
 License: Apache-2.0

{xpk-0.17.0 → xpk-0.17.2}/goldens/Workload_create.txt

@@ -35,7 +35,7 @@ docker buildx build --platform=linux/amd64 -f 4b6736a12db8ea0f78ce793fd0d4ee0c94
 docker tag dry-run-runner gcr.io/golden-project/dry-run-runner:prefix-current
 [XPK] Task: `Upload Docker Image` is implemented by the following command not running since it is a dry run. 
 docker push gcr.io/golden-project/dry-run-runner:prefix-current
-[XPK] Temp file (bd9665007cde96247406beaa13c9f40ba5636f1f75ea7561749c4472989fd14b) content: 
+[XPK] Temp file (9c33f52b12eab63cfb9ce1aba6ff74f4642cb9e40e2a0ad9517d189ee41f09a5) content: 
 apiVersion: jobset.x-k8s.io/v1alpha2
 kind: JobSet
 metadata:
@@ -65,6 +65,7 @@ spec:
           podFailurePolicy:
             rules:
             - action: FailJob
+              onPodConditions: []
               onExitCodes:
                 containerName: jax-tpu
                 operator: NotIn
@@ -145,7 +146,7 @@ spec:
               
 
 [XPK] Task: `Creating Workload` is implemented by the following command not running since it is a dry run. 
-kubectl apply -f bd9665007cde96247406beaa13c9f40ba5636f1f75ea7561749c4472989fd14b
+kubectl apply -f 9c33f52b12eab63cfb9ce1aba6ff74f4642cb9e40e2a0ad9517d189ee41f09a5
 [XPK] Task: `GKE Dashboard List` is implemented by the following command not running since it is a dry run. 
 gcloud monitoring dashboards list --project=golden-project --filter="displayName:'GKE - TPU Monitoring Dashboard'" --format="value(name)" --verbosity=error
 [XPK] Check statistics and outlier mode of GKE metrics here: https://console.cloud.google.com/monitoring/dashboards/builder/0?project=golden-project&f.rlabel.cluster_name.ClusterName=golden-cluster. To view the metric data for your workload, select golden-workload from the JobName filter on the dashboard.

{xpk-0.17.0 → xpk-0.17.2}/goldens/Workload_create_sub-slicing.txt

@@ -39,7 +39,7 @@ docker buildx build --platform=linux/amd64 -f 4b6736a12db8ea0f78ce793fd0d4ee0c94
 docker tag dry-run-runner gcr.io/golden-project/dry-run-runner:prefix-current
 [XPK] Task: `Upload Docker Image` is implemented by the following command not running since it is a dry run. 
 docker push gcr.io/golden-project/dry-run-runner:prefix-current
-[XPK] Temp file (816caa6f0023876e99f55515fda46cdbb648ff2c609f9d2af7c2354079d0e582) content: 
+[XPK] Temp file (8d5155a477cf99bc463104e0b22de0d21ee90548f51297fe429cdaa721d70a63) content: 
 apiVersion: jobset.x-k8s.io/v1alpha2
 kind: JobSet
 metadata:
@@ -69,6 +69,7 @@ spec:
           podFailurePolicy:
             rules:
             - action: FailJob
+              onPodConditions: []
               onExitCodes:
                 containerName: jax-tpu
                 operator: NotIn
@@ -150,7 +151,7 @@ spec:
               
 
 [XPK] Task: `Creating Workload` is implemented by the following command not running since it is a dry run. 
-kubectl apply -f 816caa6f0023876e99f55515fda46cdbb648ff2c609f9d2af7c2354079d0e582
+kubectl apply -f 8d5155a477cf99bc463104e0b22de0d21ee90548f51297fe429cdaa721d70a63
 [XPK] Task: `GKE Dashboard List` is implemented by the following command not running since it is a dry run. 
 gcloud monitoring dashboards list --project=golden-project --filter="displayName:'GKE - TPU Monitoring Dashboard'" --format="value(name)" --verbosity=error
 [XPK] Check statistics and outlier mode of GKE metrics here: https://console.cloud.google.com/monitoring/dashboards/builder/0?project=golden-project&f.rlabel.cluster_name.ClusterName=golden-cluster. To view the metric data for your workload, select golden-workload from the JobName filter on the dashboard.

{xpk-0.17.0 → xpk-0.17.2}/goldens/Workload_create_super-slicing.txt

@@ -39,7 +39,7 @@ docker buildx build --platform=linux/amd64 -f 4b6736a12db8ea0f78ce793fd0d4ee0c94
 docker tag dry-run-runner gcr.io/golden-project/dry-run-runner:prefix-current
 [XPK] Task: `Upload Docker Image` is implemented by the following command not running since it is a dry run. 
 docker push gcr.io/golden-project/dry-run-runner:prefix-current
-[XPK] Temp file (91fb78adbd49f9e2d6c2fec62dc461e724e79d4189af07b99e8f731bf8e2e11d) content: 
+[XPK] Temp file (5c6c507500cfbde66c80baa4f3a642c49ec3501b383057e8b68595c4121e95aa) content: 
 apiVersion: jobset.x-k8s.io/v1alpha2
 kind: JobSet
 metadata:
@@ -69,6 +69,7 @@ spec:
           podFailurePolicy:
             rules:
             - action: FailJob
+              onPodConditions: []
               onExitCodes:
                 containerName: jax-tpu
                 operator: NotIn
@@ -149,7 +150,7 @@ spec:
               
 
 [XPK] Task: `Creating Workload` is implemented by the following command not running since it is a dry run. 
-kubectl apply -f 91fb78adbd49f9e2d6c2fec62dc461e724e79d4189af07b99e8f731bf8e2e11d
+kubectl apply -f 5c6c507500cfbde66c80baa4f3a642c49ec3501b383057e8b68595c4121e95aa
 [XPK] Task: `GKE Dashboard List` is implemented by the following command not running since it is a dry run. 
 gcloud monitoring dashboards list --project=golden-project --filter="displayName:'GKE - TPU Monitoring Dashboard'" --format="value(name)" --verbosity=error
 [XPK] Check statistics and outlier mode of GKE metrics here: https://console.cloud.google.com/monitoring/dashboards/builder/0?project=golden-project&f.rlabel.cluster_name.ClusterName=golden-cluster. To view the metric data for your workload, select golden-workload from the JobName filter on the dashboard.

{xpk-0.17.0 → xpk-0.17.2}/goldens/Workload_create_with_output-manifest-file.txt

@@ -36,7 +36,7 @@ docker tag dry-run-runner gcr.io/golden-project/dry-run-runner:prefix-current
 [XPK] Task: `Upload Docker Image` is implemented by the following command not running since it is a dry run. 
 docker push gcr.io/golden-project/dry-run-runner:prefix-current
 [XPK] Workload golden-workload manifest written to /var/tmp/manifest.yaml
-[XPK] Temp file (bd9665007cde96247406beaa13c9f40ba5636f1f75ea7561749c4472989fd14b) content: 
+[XPK] Temp file (9c33f52b12eab63cfb9ce1aba6ff74f4642cb9e40e2a0ad9517d189ee41f09a5) content: 
 apiVersion: jobset.x-k8s.io/v1alpha2
 kind: JobSet
 metadata:
@@ -66,6 +66,7 @@ spec:
           podFailurePolicy:
             rules:
             - action: FailJob
+              onPodConditions: []
               onExitCodes:
                 containerName: jax-tpu
                 operator: NotIn
@@ -146,7 +147,7 @@ spec:
               
 
 [XPK] Task: `Creating Workload` is implemented by the following command not running since it is a dry run. 
-kubectl apply -f bd9665007cde96247406beaa13c9f40ba5636f1f75ea7561749c4472989fd14b
+kubectl apply -f 9c33f52b12eab63cfb9ce1aba6ff74f4642cb9e40e2a0ad9517d189ee41f09a5
 [XPK] Task: `GKE Dashboard List` is implemented by the following command not running since it is a dry run. 
 gcloud monitoring dashboards list --project=golden-project --filter="displayName:'GKE - TPU Monitoring Dashboard'" --format="value(name)" --verbosity=error
 [XPK] Check statistics and outlier mode of GKE metrics here: https://console.cloud.google.com/monitoring/dashboards/builder/0?project=golden-project&f.rlabel.cluster_name.ClusterName=golden-cluster. To view the metric data for your workload, select golden-workload from the JobName filter on the dashboard.

{xpk-0.17.0 → xpk-0.17.2}/src/xpk/commands/storage.py

@@ -23,7 +23,6 @@ from kubernetes.client.rest import ApiException
 
 from ..core import gcsfuse
 from ..core.cluster import (
-    DEFAULT_NAMESPACE,
     add_zone_and_project,
     get_cluster_network,
     setup_k8s_env,
@@ -35,12 +34,6 @@ from ..core.cluster import (
     update_cluster_with_workload_identity_if_necessary,
 )
 from ..core.filestore import FilestoreClient, get_storage_class_name
-from ..core.kjob import (
-    KJOB_API_GROUP_NAME,
-    KJOB_API_GROUP_VERSION,
-    KJOB_API_VOLUME_BUNDLE_PLURAL,
-    create_volume_bundle_instance,
-)
 from ..core.storage import (
     GCP_FILESTORE_TYPE,
     GCS_FUSE_TYPE,
@@ -98,9 +91,6 @@ def storage_create(args: Namespace) -> None:
 
     k8s_api_client = setup_k8s_env(args)
     create_storage_crds(k8s_api_client, args, manifest)
-    create_volume_bundle_instance(
-        k8s_api_client, args.name, manifest, args.readonly, args.mount_point
-    )
     # Not required for Filestore. Will be uncommented when adding GCSFuse create
     # return_code = update_cluster_with_workload_identity_if_necessary(args)
     # if return_code > 0:
@@ -214,9 +204,6 @@ def storage_attach(args: Namespace) -> None:
 
   k8s_api_client = setup_k8s_env(args)
   create_storage_crds(k8s_api_client, args, manifest)
-  create_volume_bundle_instance(
-      k8s_api_client, args.name, manifest, args.readonly, args.mount_point
-  )
 
   enable_csi_drivers_if_necessary(args)
 
@@ -332,18 +319,6 @@ def delete_storage_resources(k8s_api_client: ApiClient, storage: Storage):
         "Storage Class",
     )
 
-  delete_resource(
-      lambda name: api_instance.delete_namespaced_custom_object(
-          namespace=DEFAULT_NAMESPACE,
-          name=name,
-          group=KJOB_API_GROUP_NAME,
-          version=KJOB_API_GROUP_VERSION,
-          plural=KJOB_API_VOLUME_BUNDLE_PLURAL,
-      ),
-      storage.name,
-      "VolumeBundle",
-  )
-
   delete_resource(
       lambda name: api_instance.delete_cluster_custom_object(
           name=name,

{xpk-0.17.0 → xpk-0.17.2}/src/xpk/commands/workload.py

@@ -493,6 +493,7 @@ def workload_create(args) -> None:
           podFailurePolicy:
             rules:
             - action: FailJob
+              onPodConditions: []
               onExitCodes:
                 containerName: {get_main_container_docker_image(args, workload_system)}
                 operator: NotIn

{xpk-0.17.0 → xpk-0.17.2}/src/xpk/core/cluster.py

@@ -717,10 +717,8 @@ def get_cluster_credentials(args) -> int:
       location=location,
       dns_endpoint=True,
   )
-  if return_code != 0:
-    return return_code
 
-  if not _are_credentials_valid():
+  if return_code != 0 or not _are_credentials_valid():
     xpk_print('Detected error. Retrying without --dns-endpoint flag...')
     return_code = _get_credentials(
         project=args.project,

{xpk-0.17.0 → xpk-0.17.2}/src/xpk/parser/common.py

@@ -180,157 +180,6 @@ def add_global_arguments(custom_parser_or_group: ParserOrArgumentGroup):
   )
 
 
-def add_slurm_arguments(custom_parser_or_group: ParserOrArgumentGroup):
-  """Add Slurm job arguments to the parser.
-
-  Args:
-    custom_parser_or_group: parser or argument group to add global arguments to.
-  """
-  custom_parser_or_group.add_argument(
-      '--ignore-unknown-flags',
-      type=bool,
-      action=argparse.BooleanOptionalAction,
-      default=False,
-      help='Ignore all the unsupported flags in the bash script.',
-  )
-  custom_parser_or_group.add_argument(
-      '-a',
-      '--array',
-      type=str,
-      default=None,
-      help=(
-          'Submit a job array, multiple jobs to be executed with identical'
-          ' parameters. The indexes specification identifies what array index'
-          ' values should be used. For example, "--array=0-15" or'
-          ' "--array=0,6,16-32". Multiple values may be specified using a comma'
-          ' separated list and/or a range of values with a "-" separator. For'
-          ' example "--array=0-15%%4" will limit the number of simultaneously'
-          ' running tasks from this job array to 4. The minimum index value is'
-          ' 0. The maximum index value is 2147483647.'
-      ),
-  )
-  custom_parser_or_group.add_argument(
-      '-c',
-      '--cpus-per-task',
-      type=str,
-      default=None,
-      help='How much cpus a container inside a pod requires.',
-  )
-  custom_parser_or_group.add_argument(
-      '--gpus-per-task',
-      type=str,
-      default=None,
-      help='How much gpus a container inside a pod requires.',
-  )
-  custom_parser_or_group.add_argument(
-      '--mem',
-      type=str,
-      default=None,
-      help='How much memory a pod requires.',
-  )
-  custom_parser_or_group.add_argument(
-      '--mem-per-task',
-      type=str,
-      default=None,
-      help='How much memory a container requires.',
-  )
-  custom_parser_or_group.add_argument(
-      '--mem-per-cpu',
-      type=str,
-      default=None,
-      help=(
-          'How much memory a container requires, it multiplies the number '
-          'of requested cpus per task by mem-per-cpu.'
-      ),
-  )
-  custom_parser_or_group.add_argument(
-      '--mem-per-gpu',
-      type=str,
-      default=None,
-      help=(
-          'How much memory a container requires, it multiplies the number '
-          'of requested gpus per task by mem-per-gpu.'
-      ),
-  )
-  custom_parser_or_group.add_argument(
-      '-N',
-      '--nodes',
-      type=int,
-      default=None,
-      help='Number of pods to be used at a time.',
-  )
-  custom_parser_or_group.add_argument(
-      '-n',
-      '--ntasks',
-      type=int,
-      default=None,
-      help='Number of identical containers inside of a pod, usually 1.',
-  )
-  custom_parser_or_group.add_argument(
-      '-o',
-      '--output',
-      type=str,
-      default=None,
-      help=(
-          'Where to redirect the standard output stream of a task. If not'
-          ' passed it proceeds to stdout, and is available via kubectl logs.'
-      ),
-  )
-  custom_parser_or_group.add_argument(
-      '-e',
-      '--error',
-      type=str,
-      default=None,
-      help=(
-          'Where to redirect std error stream of a task. If not passed it'
-          ' proceeds to stdout, and is available via kubectl logs.'
-      ),
-  )
-  custom_parser_or_group.add_argument(
-      '--input',
-      type=str,
-      default=None,
-      help='What to pipe into the script.',
-  )
-  custom_parser_or_group.add_argument(
-      '-J',
-      '--job-name',
-      type=str,
-      default=None,
-      help='What is the job name.',
-  )
-  custom_parser_or_group.add_argument(
-      '-D',
-      '--chdir',
-      type=str,
-      default=None,
-      help='Change directory before executing the script.',
-  )
-  custom_parser_or_group.add_argument(
-      '-t',
-      '--time',
-      type=str,
-      default=None,
-      help=(
-          'Set a limit on the total run time of the job. '
-          'A time limit of zero requests that no time limit be imposed. '
-          'Acceptable time formats include "minutes", "minutes:seconds", '
-          '"hours:minutes:seconds", "days-hours", "days-hours:minutes" '
-          'and "days-hours:minutes:seconds".'
-      ),
-  )
-  custom_parser_or_group.add_argument(
-      '--priority',
-      type=str,
-      default='medium',
-      choices=['very-low', 'low', 'medium', 'high', 'very-high'],
-      help=(
-          'A priority, one of `very-low`, `low`, `medium`, `high` or'
-          ' `very-high`. Defaults to `medium`.'
-      ),
-  )
-
-
 def add_tpu_type_argument(
     custom_parser_or_group: ParserOrArgumentGroup,
     required: bool = False,

{xpk-0.17.0 → xpk-0.17.2/src/xpk.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: xpk
-Version: 0.17.0
+Version: 0.17.2
 Summary: xpk helps Cloud developers to orchestrate training jobs on accelerators on GKE.
 Author-email: XPK team <xpk-code-reviewers@google.com>
 License: Apache-2.0

{xpk-0.17.0 → xpk-0.17.2}/src/xpk.egg-info/SOURCES.txt

@@ -182,7 +182,6 @@ src/xpk/core/gcloud_context_test.py
 src/xpk/core/gcluster_manager.py
 src/xpk/core/gcsfuse.py
 src/xpk/core/jobset.py
-src/xpk/core/kjob.py
 src/xpk/core/kueue_manager.py
 src/xpk/core/kueue_manager_test.py
 src/xpk/core/monitoring.py
@@ -262,7 +261,6 @@ src/xpk/templates/kueue_sub_slicing_topology.yaml.j2
 src/xpk/templates/kueue_super_slicing_topology.yaml.j2
 src/xpk/templates/mtc-cpc.yaml
 src/xpk/templates/storage.yaml
-src/xpk/templates/volume_bundle.yaml
 src/xpk/utils/__init__.py
 src/xpk/utils/console.py
 src/xpk/utils/console_test.py