Add telemetry and dog fooding fixes (#248)

* add telemetry to init experience, remove duplicate code in init_constants

* add filter for deprecation warning, fix hyp --version

* change default instance group name for instance group settings

Author: mollyheamazon

examples/cluster_management/cluster_creation_sdk_experience.ipynb

@@ -40,7 +40,9 @@
   },
   {
    "cell_type": "code",
+   "execution_count": null,
    "metadata": {},
+   "outputs": [],
    "source": [
     "import uuid\n",
     "import time\n",
@@ -83,7 +85,7 @@
     "    instance_group_settings=[\n",
     "        {\n",
     "            \"InstanceCount\": 1,\n",
-    "            \"InstanceGroupName\": \"controller-group\",\n",
+    "            \"InstanceGroupName\": \"default\",\n",
     "            \"InstanceType\": \"ml.t3.medium\",\n",
     "            \"TargetAvailabilityZoneId\": \"use2-az2\",\n",
     "            \"ThreadsPerCore\": 1,\n",
@@ -96,9 +98,7 @@
     "\n",
     "print(f\"Initialized cluster stack with prefix: {resource_prefix}\")\n",
     "print(f\"Cluster name: {cluster_stack.hyperpod_cluster_name}\")"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
    "cell_type": "markdown",
@@ -117,7 +117,9 @@
   },
   {
    "cell_type": "code",
+   "execution_count": null,
    "metadata": {},
+   "outputs": [],
    "source": [
     "# Configure cluster with custom tags (equivalent to hyp configure --tags)\n",
     "cluster_tags = [\n",
@@ -143,9 +145,7 @@
     "\n",
     "print(f\"\\nNode recovery: {cluster_stack.node_recovery}\")\n",
     "print(f\"FSx storage capacity: {cluster_stack.storage_capacity} GiB\")"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
    "cell_type": "markdown",
@@ -158,7 +158,9 @@
   },
   {
    "cell_type": "code",
+   "execution_count": null,
    "metadata": {},
+   "outputs": [],
    "source": [
     "# Display current configuration details\n",
     "print(\"=== Cluster Configuration ===\")\n",
@@ -176,9 +178,7 @@
     "print(f\"  EKS Stack: {cluster_stack.create_eks_cluster_stack}\")\n",
     "print(f\"  HyperPod Stack: {cluster_stack.create_hyperpod_cluster_stack}\")\n",
     "print(f\"  FSx Stack: {cluster_stack.create_fsx_stack}\")"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
    "cell_type": "markdown",
@@ -201,7 +201,9 @@
   },
   {
    "cell_type": "code",
+   "execution_count": null,
    "metadata": {},
+   "outputs": [],
    "source": [
     "# Create the HyperPod cluster (equivalent to hyp create)\n",
     "try:\n",
@@ -225,9 +227,7 @@
     "except Exception as e:\n",
     "    print(f\"\\n❌ Cluster creation failed: {str(e)}\")\n",
     "    raise"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
    "cell_type": "markdown",
@@ -240,7 +240,9 @@
   },
   {
    "cell_type": "code",
+   "execution_count": null,
    "metadata": {},
+   "outputs": [],
    "source": [
     "# Monitor cluster creation progress\n",
     "def monitor_cluster_creation(stack_name, max_checks=30, interval=120):\n",
@@ -278,9 +280,7 @@
     "# Start monitoring (uncomment when cluster creation is initiated)\n",
     "# final_status = monitor_cluster_creation(stack_name, max_checks=5, interval=30)\n",
     "print(\"Monitoring function ready. Uncomment to start monitoring after cluster creation.\")"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
    "cell_type": "markdown",
@@ -300,7 +300,9 @@
   },
   {
    "cell_type": "code",
+   "execution_count": null,
    "metadata": {},
+   "outputs": [],
    "source": [
     "# Get detailed information about the cluster stack (equivalent to hyp describe cluster-stack)\n",
     "def describe_cluster_stack(stack_name, region=\"us-east-2\"):\n",
@@ -347,9 +349,7 @@
     "# Describe the cluster stack (uncomment when stack exists)\n",
     "# describe_cluster_stack(stack_name)\n",
     "print(\"Describe function ready. Use after cluster creation is complete.\")"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
    "cell_type": "markdown",
@@ -368,7 +368,9 @@
   },
   {
    "cell_type": "code",
+   "execution_count": null,
    "metadata": {},
+   "outputs": [],
    "source": [
     "# List all cluster stacks (equivalent to hyp list cluster-stack)\n",
     "def list_cluster_stacks(region=\"us-east-2\"):\n",
@@ -416,9 +418,7 @@
     "        print(f\"\\n=== HyperPod Stacks ({len(hyperpod_stacks)}) ===\")\n",
     "        for stack in hyperpod_stacks:\n",
     "            print(f\"  - {stack['StackName']} ({stack['StackStatus']})\")"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
    "cell_type": "markdown",
@@ -439,7 +439,9 @@
   },
   {
    "cell_type": "code",
+   "execution_count": null,
    "metadata": {},
+   "outputs": [],
    "source": [
     "# Update cluster configuration using sagemaker-core Cluster class\n",
     "def update_cluster(cluster_name, region=\"us-east-2\"):\n",
@@ -524,9 +526,7 @@
     "# scaled_cluster = scale_instance_group(cluster_name, \"controller-group\", 2)\n",
     "\n",
     "print(\"Update functions ready. Use after cluster creation is complete.\")"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
    "cell_type": "markdown",
@@ -539,7 +539,9 @@
   },
   {
    "cell_type": "code",
+   "execution_count": null,
    "metadata": {},
+   "outputs": [],
    "source": [
     "# Comprehensive cluster health check\n",
     "def check_cluster_health(cluster_name, region=\"us-east-2\"):\n",
@@ -596,9 +598,7 @@
     "# cluster_health = check_cluster_health(cluster_name)\n",
     "\n",
     "print(\"Health check function ready. Use after cluster creation is complete.\")"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
    "cell_type": "markdown",

hyperpod-cluster-stack-template/hyperpod_cluster_stack_template/creation_template.yaml

@@ -320,7 +320,7 @@ Parameters:
   InstanceGroupSettings1:
     Type: String
     Default: >-
-      [{"InstanceCount":1,"InstanceGroupName":"controller-group","InstanceType":"ml.t3.medium","TargetAvailabilityZoneId":"use2-az2","ThreadsPerCore":1,"InstanceStorageConfigs":[{"EbsVolumeConfig":{"VolumeSizeInGB":500}}]}]
+      [{"InstanceCount":1,"InstanceGroupName":"default","InstanceType":"ml.t3.medium","TargetAvailabilityZoneId":"use2-az2","ThreadsPerCore":1,"InstanceStorageConfigs":[{"EbsVolumeConfig":{"VolumeSizeInGB":500}}]}]
     Description: JSON array string containing instance group configurations.
   RigS3BucketName:
     Type: String

hyperpod-cluster-stack-template/hyperpod_cluster_stack_template/v1_0/model.py

@@ -17,7 +17,7 @@ class ClusterStackBase(BaseModel):
     helm_release: Optional[str] = Field("dependencies", description="The name used for Helm chart release")
     node_provisioning_mode: Optional[str] = Field("Continuous", description="Enable or disable the continuous provisioning mode. Valid values: \"Continuous\" or leave empty")
     node_recovery: Optional[str] = Field("Automatic", description="Specifies whether to enable or disable the automatic node recovery feature. Valid values: \"Automatic\", \"None\"")
-    instance_group_settings: Union[List[Any], None] = Field([{"InstanceCount":1,"InstanceGroupName":"controller-group","InstanceType":"ml.t3.medium","TargetAvailabilityZoneId":"use2-az2","ThreadsPerCore":1,"InstanceStorageConfigs":[{"EbsVolumeConfig":{"VolumeSizeInGB":500}}]}], description="List of string containing instance group configurations")
+    instance_group_settings: Union[List[Any], None] = Field([{"InstanceCount":1,"InstanceGroupName":"default","InstanceType":"ml.t3.medium","TargetAvailabilityZoneId":"use2-az2","ThreadsPerCore":1,"InstanceStorageConfigs":[{"EbsVolumeConfig":{"VolumeSizeInGB":500}}]}], description="List of string containing instance group configurations")
     rig_settings: Union[List[Any], None] = Field(None, description="List of string containing restricted instance group configurations")
     rig_s3_bucket_name: Optional[str] = Field(None, description="The name of the S3 bucket used to store the RIG resources")
     tags: Union[List[Any], None] = Field(None, description="Custom tags for managing the SageMaker HyperPod cluster as an AWS resource")

hyperpod-cluster-stack-template/hyperpod_cluster_stack_template/v1_0/schema.json

@@ -181,7 +181,7 @@
       "default": [
         {
           "InstanceCount": 1,
-          "InstanceGroupName": "controller-group",
+          "InstanceGroupName": "default",
           "InstanceType": "ml.t3.medium",
           "TargetAvailabilityZoneId": "use2-az2",
           "ThreadsPerCore": 1,

src/sagemaker/hyperpod/cli/__init__.py

@@ -0,0 +1,9 @@
+import warnings
+# Reset warnings and show all except Pydantic serialization warnings
+warnings.resetwarnings()
+warnings.simplefilter("always")
+# Suppress specific Pydantic serialization warnings globally (this is ignored due to customized parsing logic)
+warnings.filterwarnings("ignore", message=".*PydanticSerializationUnexpectedValue.*", category=UserWarning)
+warnings.filterwarnings("ignore", message=".*serializer.*", category=UserWarning, module="pydantic")
+# Suppress kubernetes urllib3 deprecation warning (this is internal dependencies)
+warnings.filterwarnings("ignore", message=".*HTTPResponse.getheaders.*", category=DeprecationWarning, module="kubernetes")

src/sagemaker/hyperpod/cli/commands/init.py

@@ -25,11 +25,16 @@
     get_default_version_for_template
 )
 from sagemaker.hyperpod.common.utils import get_aws_default_region
+from sagemaker.hyperpod.common.telemetry.telemetry_logging import (
+    _hyperpod_telemetry_emitter,
+)
+from sagemaker.hyperpod.common.telemetry.constants import Feature
 
 @click.command("init")
 @click.argument("template", type=click.Choice(list(TEMPLATES.keys())))
 @click.argument("directory", type=click.Path(file_okay=False), default=".")
 @click.option("--version", "-v", default=None, help="Schema version")
+@_hyperpod_telemetry_emitter(Feature.HYPERPOD_CLI, "init_template_cli")
 def init(
     template: str,
     directory: str,
@@ -144,6 +149,7 @@ def init(
 
 
 @click.command("reset")
+@_hyperpod_telemetry_emitter(Feature.HYPERPOD_CLI, "init_reset_cli")
 def reset():
     """
     Reset the current directory's config.yaml to an "empty" scaffold:
@@ -176,6 +182,7 @@ def reset():
 @click.command("configure")
 @generate_click_command()
 @click.pass_context
+@_hyperpod_telemetry_emitter(Feature.HYPERPOD_CLI, "init_configure_cli")
 def configure(ctx, model_config):
     """
     Update any subset of fields in ./config.yaml by passing --<field> flags.
@@ -253,6 +260,7 @@ def configure(ctx, model_config):
 
 
 @click.command("validate")
+@_hyperpod_telemetry_emitter(Feature.HYPERPOD_CLI, "init_validate_cli")
 def validate():
     """
     Validate this directory's config.yaml against the appropriate schema.
@@ -263,6 +271,7 @@ def validate():
 
 @click.command(name="_default_create")
 @click.option("--region", "-r", default=None, help="Region to create cluster stack for, default to your region in aws configure. Not available for other templates.")
+@_hyperpod_telemetry_emitter(Feature.HYPERPOD_CLI, "init_create_cli")
 def _default_create(region):
     """
     Validate configuration and render template files for deployment.

src/sagemaker/hyperpod/cli/constants/init_constants.py

@@ -71,37 +71,6 @@ def version_key(v):
     'hyp-pytorch-job.volume': _get_handler_from_latest_template("hyp-pytorch-job", "VOLUME_TYPE_HANDLER"),
 }
 
-
-def _get_handler_from_latest_template(template_name, handler_name):
-    """Dynamically import handler from the latest version of a template"""
-    try:
-        template_info = TEMPLATES[template_name]
-        registry = template_info["registry"]
-
-        # Get latest version using same logic as _get_latest_class
-        available_versions = list(registry.keys())
-        def version_key(v):
-            try:
-                return tuple(map(int, v.split('.')))
-            except ValueError:
-                return (0, 0)
-
-        latest_version = max(available_versions, key=version_key)
-        latest_model = registry[latest_version]
-        
-        # Get handler from module
-        module = sys.modules[latest_model.__module__]
-        return getattr(module, handler_name)
-    except (ImportError, AttributeError):
-        return None
-
-
-# Template.field to handler mapping - avoids conflicts and works reliably
-SPECIAL_FIELD_HANDLERS = {
-    'hyp-pytorch-job.volume': _get_handler_from_latest_template("hyp-pytorch-job", "VOLUME_TYPE_HANDLER"),
-}
-
-
 USAGE_GUIDE_TEXT_CFN = """# SageMaker HyperPod CLI - Initialization Workflow
 
 This document explains the initialization workflow and related commands for the SageMaker HyperPod CLI.

src/sagemaker/hyperpod/cli/hyp_cli.py

@@ -1,11 +1,3 @@
-import warnings
-# Reset warnings and show all except Pydantic serialization warnings
-warnings.resetwarnings()
-warnings.simplefilter("always")
-# Suppress specific Pydantic serialization warnings globally
-warnings.filterwarnings("ignore", message=".*PydanticSerializationUnexpectedValue.*", category=UserWarning)
-warnings.filterwarnings("ignore", message=".*serializer.*", category=UserWarning, module="pydantic")
-
 import click
 import yaml
 import json
@@ -56,7 +48,6 @@
 )
 
 
-@click.group(context_settings={'max_content_width': 200})
 def get_package_version(package_name):
     try:
         return version(package_name)

src/sagemaker/hyperpod/common/telemetry/telemetry_logging.py

@@ -149,12 +149,31 @@ def _hyperpod_telemetry_emitter(feature: str, func_name: str):
     def decorator(func):
         @functools.wraps(func)
         def wrapper(*args, **kwargs):
+            import inspect
+            sig = inspect.signature(func)
+            bound_args = sig.bind(*args, **kwargs)
+            bound_args.apply_defaults()
+            
+            # Get template value and create template-specific event name
+            template = bound_args.arguments.get('template')
+            if template:
+                event_name = f"{func_name}_{template.replace('-', '_')}"
+            else:
+                event_name = func_name
+            
             extra = (
-                f"{func_name}"
+                f"{event_name}"
                 f"&x-sdkVersion={SDK_VERSION}"
                 f"&x-env={PYTHON_VERSION}"
                 f"&x-sys={OS_NAME_VERSION}"
             )
+
+            # Add template and version to extra
+            if template:
+                extra += f"&x-template={template}"
+            if 'version' in bound_args.arguments and bound_args.arguments['version']:
+                extra += f"&x-version={bound_args.arguments['version']}"
+            
             start = perf_counter()
             try:
                 result = func(*args, **kwargs)