Workspace Policies

This topic describes APIs to list, retrieve, and update workspace policies.

List Workspace Policies

Use this API to get all workspace policies that are defined in a workspace.

Method and URL

/api/v1/workspace-policies

Parameters

Response Parameters

Parameter	Parameter Type	Description
workspaceName	String	The name of the workspace associated with the policy.
PriorityRange	String	The priority range configuration for the workspace.
maxGPRs	Integer	The maximum number of GPU pod requests allowed.
maxExitDurationPerGPR	String	The maximum exit duration per GPU pod request.
requeueOnFailure	Boolean	This is the parameter to requeue requests on failure.
enableAutoEviction	Boolean	This is the parameter to enable auto eviction for the workspace.
globalGPUConstraints	Object	The global GPU constraints configuration.
globalGPUConstraints.gpuShapes	Array of Strings]	The list of allowed GPU shapes.
globalGPUConstraints.maxGPUPerGpr	Integer	The maximum GPUs per GPU Pod Request.
globalGPUConstraints.maxMemoryPerGpr	Integer	The maximum memory per GPU Pod Request (in bytes or MB).

Example Request

GET /api/v1/workspace-policies

Example Response

Example Responses: Success

{
  "statusCode": 200,
  "status": "OK",
  "message": "Success",
  "data": [
    {
      "workspaceName": "workspace-1",
      "PriorityRange": "100-200",
      "maxGPRs": 10,
      "maxExitDurationPerGPR": "1h",
      "requeueOnFailure": true,
      "enableAutoEviction": false,
      "globalGPUConstraints": {
        "gpuShapes": ["NVIDIA-A100", "NVIDIA-V100"],
        "maxGPUPerGpr": 4,
        "maxMemoryPerGpr": 64
      }
    },
    {
      "workspaceName": "workspace-2",
      "PriorityRange": "200-300",
      "maxGPRs": 5,
      "maxExitDurationPerGPR": "2h",
      "requeueOnFailure": false,
      "enableAutoEviction": true,
      "globalGPUConstraints": {
        "gpuShapes": ["NVIDIA-T4"],
        "maxGPUPerGpr": 2,
        "maxMemoryPerGpr": 32
      }
    }
  ]
}

Example: Error

{
  "statusCode": 500,
  "status": "Error",
  "message": "Internal Server Error",
  "data": null,
  "error": {
    "errorKey": "UPSTREAM_REQUEST_ERROR",
    "message": "Failed to fetch workspace policies",
    "data": "{\"error\":{\"Message\":\"Error retrieving workspace policies\",\"Status\":500,\"DetailedError\":{\"Errormessage\":\"API gateway timeout\",\"Reason\":\"InternalError\",\"Details\":\"\"},\"StatusCode\":500}}"
  }
}

List Workspace Policy

Use this API to get a required Workspace Policy.

Method and URL

GET /api/v1/workspace-policies/{policyName}

Change {policyName} to the actual name of the policy you want to retrieve.

Parameters

Response Parameters

Parameter	Parameter Type	Description
workspaceName	String	The name of the workspace associated with the policy.
PriorityRange	String	The priority range configuration for the workspace.
maxGPRs	Integer	The maximum number of GPU Pod Requests allowed.
maxExitDurationPerGPR	String	The maximum exit duration per GPU Pod Request.
requeueOnFailure	Boolean	This parameter is to enable requeue requests on failure.
enableAutoEviction	Boolean	This parameter is to enable auto eviction for the workspace.
globalGPUConstraints	Object	The global GPU constraints configuration.
globalGPUConstraints.gpuShapes	Array[string]	The list of allowed GPU shapes.
globalGPUConstraints.maxGPUPerGpr	Integer	The maximum GPUs per GPU Pod Request.
globalGPUConstraints.maxMemoryPerGpr	Integer	The maximum memory per GPU Pod Request (in bytes or MB).

Example Request

GET /api/v1/workspace-policies/workspace-1

Example Responses

Example Response: Success

{
  "statusCode": 200,
  "status": "OK",
  "message": "Success",
  "data": {
    "workspaceName": "workspace-1",
    "PriorityRange": "100-200",
    "maxGPRs": 10,
    "maxExitDurationPerGPR": "1h",
    "requeueOnFailure": true,
    "enableAutoEviction": false,
    "globalGPUConstraints": {
      "gpuShapes": ["NVIDIA-A100", "NVIDIA-V100"],
      "maxGPUPerGpr": 4,
      "maxMemoryPerGpr": 64
    }
  }
}

Example Response: Error

{
  "statusCode": 404,
  "status": "NOT_FOUND",
  "message": "Workspace policy not found",
  "data": null,
  "error": {
    "errorKey": "UPSTREAM_REQUEST_ERROR",
    "message": "Workspace policy 'workspace-1' not found",
    "data": "{\"error\":{\"Message\":\"Policy not found\",\"Status\":404,\"DetailedError\":{\"Errormessage\":\"No policy found with name workspace-1\",\"Reason\":\"NotFound\",\"Details\":\"\"},\"StatusCode\":404}}"
  }
}

Update Workspace Policy

Use this API to update a Workspace Policy.

Method and URL

PUT /api/v1/workspace-policies/{policyName}

Change {policyName} to the actual name of the policy you want to update.

Parameters

Parameter	Parameter Type	Description	Required
policyName	String	The name of the workspace policy you want to update.	Mandatory
priorityRange	String	The priority range configuration for the workspace.	Optional
maxGPRs	Integer	The maximum number of GPU Pod Requests allowed.	Optional
maxExitDurationPerGPR	String	The maximum exit duration per GPU Pod Request.	Optional
enforceIdleTimeOut	Boolean	The parameter to enforce idle timeout for GPU Pod Requests.	Optional
requeueOnFailure	Boolean	The parameter to requeue requests on failure.	Optional
enableAutoEviction	Boolean	The parameter to enable auto eviction for the workspace.	Optional
gpuShapes	Array[string]	The list of allowed GPU shapes.	Optional
maxGpuPerGpr	Integer	The maximum GPUs per GPU Pod Request.	Optional
maxMemoryPerGpr	Integer	The maximum memory per GPU Pod Request (in bytes or MB).	Optional

Example Request

{
  "priorityRange": "150-250",
  "maxGPRs": 15,
  "requeueOnFailure": false,
  "gpuShapes": ["NVIDIA-A100"],
  "maxGpuPerGpr": 8,
  "maxMemoryPerGpr": 128
}

Example Responses

Example Response: Success

{
  "statusCode": 200,
  "status": "OK",
  "message": "Success",
  "data": {
    "workspaceName": "workspace-1",
    "PriorityRange": "150-250",
    "maxGPRs": 15,
    "maxExitDurationPerGPR": "1h",
    "requeueOnFailure": false,
    "enableAutoEviction": false,
    "globalGPUConstraints": {
      "gpuShapes": ["NVIDIA-A100"],
      "maxGPUPerGpr": 8,
      "maxMemoryPerGpr": 128
    }
  }
}

Example Response: Bad Request

{
  "statusCode": 400,
  "status": "INVALID_REQUEST",
  "message": "At least one field must be provided for update",
  "data": null,
  "error": {
    "errorKey": "INVALID_REQUEST",
    "message": "At least one field must be provided for update",
    "data": null
  }
}

Example Response: Not Found

{
  "statusCode": 404,
  "status": "NOT_FOUND",
  "message": "Workspace policy not found",
  "data": null,
  "error": {
    "errorKey": "UPSTREAM_REQUEST_ERROR",
    "message": "Workspace policy 'workspace-1' not found",
    "data": "{\"error\":{\"Message\":\"Policy not found\",\"Status\":404,\"DetailedError\":{\"Errormessage\":\"No policy found with name workspace-1\",\"Reason\":\"NotFound\",\"Details\":\"\"},\"StatusCode\":404}}"
  }
}

Example Response: Internal Server Error

{
  "statusCode": 500,
  "status": "Error",
  "message": "Internal Server Error",
  "data": null,
  "error": {
    "errorKey": "UPSTREAM_REQUEST_ERROR",
    "message": "Failed to update workspace policy",
    "data": "{\"error\":{\"Message\":\"Error updating workspace policy\",\"Status\":500,\"DetailedError\":{\"Errormessage\":\"API gateway timeout\",\"Reason\":\"InternalError\",\"Details\":\"\"},\"StatusCode\":500}}"
  }
}