Merge branch 'aws:main' into main-feature-get-op-logs

Author: rsareddy0329

README.md

@@ -54,24 +54,13 @@ SageMaker HyperPod CLI currently supports start training job with:
 
 1. Make sure that your local python version is 3.8, 3.9, 3.10 or 3.11.
 
-1. Install ```helm```.
-
-    The SageMaker Hyperpod CLI uses Helm to start training jobs. See also the [Helm installation guide](https://helm.sh/docs/intro/install/).
-
-    ```
-    curl -fsSL -o get_helm.sh https://raw.githubusercontent.com/helm/helm/main/scripts/get-helm-3
-    chmod 700 get_helm.sh
-    ./get_helm.sh
-    rm -f ./get_helm.sh  
-    ```
-
-1. Clone and install the sagemaker-hyperpod-cli package.
+2. Install the sagemaker-hyperpod-cli package.
 
     ```
     pip install sagemaker-hyperpod
     ```
 
-1. Verify if the installation succeeded by running the following command.
+3. Verify if the installation succeeded by running the following command.
 
     ```
     hyp --help
@@ -207,7 +196,7 @@ hyp invoke hyp-jumpstart-endpoint \
 
 ```
 hyp list hyp-jumpstart-endpoint
-hyp get hyp-jumpstart-endpoint --name endpoint-jumpstart
+hyp describe hyp-jumpstart-endpoint --name endpoint-jumpstart
 ```
 
 #### Creating a Custom Inference Endpoint 

doc/cli_training.md

@@ -40,7 +40,7 @@ hyp create hyp-pytorch-job [OPTIONS]
 - `--tasks-per-node INTEGER`: Number of tasks per node (minimum: 1)
 - `--label-selector OBJECT`: Node label selector as key-value pairs
 - `--deep-health-check-passed-nodes-only BOOLEAN`: Schedule pods only on nodes that passed deep health check (default: false)
-- `--scheduler-type TEXT`: Scheduler type
+- `--scheduler-type TEXT`: If specified, training job pod will be dispatched by specified scheduler. If not specified, the pod will be dispatched by default scheduler.
 - `--queue-name TEXT`: Queue name for job scheduling (1-63 characters, alphanumeric with hyphens)
 - `--priority TEXT`: Priority class for job scheduling
 - `--max-retry INTEGER`: Maximum number of job retries (minimum: 0)

helm_chart/HyperPodHelmChart/Chart.yaml

@@ -24,6 +24,10 @@ version: 0.1.0
 appVersion: "1.16.0"
 
 dependencies:
+  - name: cert-manager
+    version: "v1.18.2"
+    repository: oci://quay.io/jetstack/charts
+    condition: cert-manager.enabled
   - name: training-operators
     version: "0.1.0"
     repository: "file://charts/training-operators"

helm_chart/HyperPodHelmChart/values.yaml

@@ -115,6 +115,15 @@ namespace:
   create: true
   name: aws-hyperpod
 
+cert-manager:
+  enabled: true
+  namespace: cert-manager
+  global:
+    leaderElection:
+      namespace: cert-manager
+  crds:
+    enabled: true
+
 mlflow:
   enabled: false
 

helm_chart/readme.md

@@ -33,6 +33,7 @@ More information about orchestration features for cluster admins [here](https://
 | [Kubeflow Training Operator](https://www.kubeflow.org/docs/components/trainer/legacy-v1/overview/)            | Installs operators for managing various machine learning training jobs, such as TensorFlow, PyTorch, and MXNet, providing native Kubernetes support for distributed training workloads. |              | Yes               |
 | HyperPod patching            | Deploys the RBAC and controller resources needed for orchestrating rolling updates and patching workflows in SageMaker HyperPod clusters. Includes pod eviction and node monitoring.    | HyperPod Resiliency             | Yes               |
 | hyperpod-inference-operator  | Installs the HyperPod Inference Operator and its dependencies to the cluster, allowing cluster deployment and inferencing of JumpStart, s3-hosted, and FSx-hosted models                | No                | 
+| [cert-manager](https://github.com/cert-manager/cert-manager)                | Automatically provisions and manages TLS certificates in Kubernetes clusters. Provides certificate lifecycle management including issuance, renewal, and revocation for secure communications. | [Hyperpod training operator](https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-eks-operator.html)           | Yes                |
 
 > **_Note_** The `mpijob` scheme is disabled in the Training Operator helm chart to avoid conflicting with the MPI Operator. 
 
@@ -48,6 +49,20 @@ storage:
   enabled: true
 ```
 
+To enable cert-manager for TLS certificate management, pass in `--set cert-manager.enabled=true` when installing or upgrading the main chart or set the following in the values.yaml file:
+```
+cert-manager:
+  enabled: true
+  namespace: cert-manager
+  global:
+    leaderElection:
+      namespace: cert-manager
+  crds:
+    enabled: true  
+```
+namespace specifies which name space cert-manager should be installed
+
+
 ---
 
 The following plugins are only required for HyperPod Resiliency if you are using the following supported devices, such as GPU/Neuron instances, unless you install these plugins on your own. 

hyperpod-custom-inference-template/hyperpod_custom_inference_template/v1_0/model.py

@@ -10,7 +10,7 @@
 # distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF
 # ANY KIND, either express or implied. See the License for the specific
 # language governing permissions and limitations under the License.
-from pydantic import BaseModel, Field
+from pydantic import BaseModel, Field, model_validator, ConfigDict
 from typing import Optional, List, Dict, Union, Literal
 
 from sagemaker.hyperpod.inference.config.hp_endpoint_config import (
@@ -31,9 +31,19 @@
 from sagemaker.hyperpod.inference.hp_endpoint import HPEndpoint
 
 class FlatHPEndpoint(BaseModel):
+    model_config = ConfigDict(extra="forbid")
+
+    metadata_name: Optional[str]  = Field(
+        None,
+        alias="metadata_name",
+        description="Name of the jumpstart endpoint object",
+        max_length=63,
+        pattern=r"^[a-zA-Z0-9](-*[a-zA-Z0-9]){0,62}$",
+    )
+
     # endpoint_name
     endpoint_name: Optional[str] = Field(
-        "",
+        None,
         alias="endpoint_name",
         description="Name of SageMaker endpoint; empty string means no creation",
         max_length=63,
@@ -130,7 +140,7 @@ class FlatHPEndpoint(BaseModel):
         description="FSX File System DNS Name",
     )
     fsx_file_system_id: Optional[str] = Field(
-        ...,  
+        None,  
         alias="fsx_file_system_id",
         description="FSX File System ID",
     )
@@ -142,12 +152,12 @@ class FlatHPEndpoint(BaseModel):
 
     # S3Storage
     s3_bucket_name: Optional[str] = Field(
-        ..., 
+        None, 
         alias="s3_bucket_name",
         description="S3 bucket location",
     )
     s3_region: Optional[str] = Field(
-        ..., 
+        None, 
         alias="s3_region",
         description="S3 bucket region",
     )
@@ -229,12 +239,22 @@ class FlatHPEndpoint(BaseModel):
     invocation_endpoint: Optional[str] = Field(
         default="invocations",
         description=(
-            "The invocation endpoint of the model server. "
-            "http://<host>:<port>/ would be pre-populated based on the other fields. "
+            "The invocation endpoint of the model server. http://<host>:<port>/ would be pre-populated based on the other fields. "
             "Please fill in the path after http://<host>:<port>/ specific to your model server.",
         )
     )
-
+    
+    @model_validator(mode='after')
+    def validate_model_source_config(self):
+        """Validate that required fields are provided based on model_source_type"""
+        if self.model_source_type == "s3":
+            if not self.s3_bucket_name or not self.s3_region:
+                raise ValueError("s3_bucket_name and s3_region are required when model_source_type is 's3'")
+        elif self.model_source_type == "fsx":
+            if not self.fsx_file_system_id:
+                raise ValueError("fsx_file_system_id is required when model_source_type is 'fsx'")
+        return self
+    
     def to_domain(self) -> HPEndpoint:
         env_vars = None
         if self.env: