> This page is for Self-Hosted, version v3.0 (default). > For other versions, use one of these documentation indexes: > - v3.0 (default): https://docs.nvidia.com/run-ai/self-hosted/v3.0/llms.txt > - v2.26: https://docs.nvidia.com/run-ai/self-hosted/v2.26/llms.txt > - v2.25: https://docs.nvidia.com/run-ai/self-hosted/v2.25/llms.txt > - v2.24: https://docs.nvidia.com/run-ai/self-hosted/v2.24/llms.txt > For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://docs.nvidia.com/run-ai/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/run-ai/_mcp/server. # List S3 assets. GET http://localhost:8080/api/v1/asset/datasource/s3 Retrieve a list of S3 compatible datasource assets. Reference: https://docs.nvidia.com/run-ai/self-hosted/reference/rest-api/endpoints/workload-assets/s-3/list-s-3-assets ## Authentication - `Authorization` header (bearer token, required) — Bearer authentication ## Request ### Query parameters - `name` (string, optional) — Filter results by name. - `scope` (string, optional) — Filters results by scope. Returns only assets that belong to the specified scope. Mutually exclusive with includeDescendants. Valid values: tenant, cluster, department, project. - `projectId` (integer, optional) — Filter results by project id. If scope filter is project, only assets from the specific project will be included in the response. Otherwise, the response will include project, department, cluster, tenant and system assets. - `departmentId` (string, optional) — Filter results by department id. If scope filter is department, only assets from the specific department will be included in the response. Otherwise, the response will include department, cluster, tenant and system assets. - `clusterId` (string, optional) — Filter results by Universally Unique Identifier (UUID) of the cluster. If scope filter is cluster, only assets from the specific cluster will be included in the response. Otherwise, the response will include cluster, tenant and system assets. - `usageInfo` (boolean, optional) — Whether the query should include asset usage information as part of the response. - `complyToProject` (integer, optional) — Include workload creation compliance information of an asset, for a given project, as part of the response. To check compliance, you need to provide both project id and workload type. - `complyToWorkloadType` (enum, optional) — Include workload creation compliance information of an asset, for a given workload type, as part of the response. To check compliance, you need to provide both project id and workload type. - Allowed values: `Workspace`, `Training`, `Distributed`, `Inference`, `DistributedInference` - `statusInfo` (boolean, optional) — Whether the query should include asset status information as part of the response. - `assetIds` (string, optional) — Filter results by the ids of the assets. Provided value should be a comma separated string of UUIDs. - `complyToReplicaType` (enum, optional) — Include workload creation compliance information of an asset, for a given replica type, as part of the response. To check compliance, you need to provide both project id and workload type. For distributed, replica type should be provided as well. - Allowed values: `Worker`, `Master`, `Leader` ## Response ### 200 Request completed successfully. - `entries` (list of S3Asset, required) ## Errors ### 401 Unauthorized Error Unauthorized - `code` (integer, required) - `message` (string, required) - `details` (string, optional) ### 403 Forbidden Error Forbidden - `code` (integer, required) - `message` (string, required) - `details` (string, optional) ### 503 Service Unavailable Error unexpected error - `code` (integer, required) - `message` (string, required) - `details` (string, optional) ## Types ### S3Asset - `meta` (AssetMeta, required) - `spec` (S3AssetSpec, required) - `usedBy` (AssetUsageInfo, optional) — Details about resources that use the asset. - `usageTimes` (UsageTimesInfo, optional) — Details about times the asset has been used by workloads, etc. - `compliance` (ComplianceInfo, optional) — Compliance of an asset with creation of workloads in a given project. - `status` (AssetClusterStatusInfo, optional) ### AssetMeta - `name` (string, required) — The name of the asset. - `scope` (enum, required) — The scope in which an asset can be used. The highest scope, system, is intended for internal purposes only. - Allowed values: `system`, `tenant`, `cluster`, `department`, `project` - `id` (string, required) — Unique identifier of the asset. - `kind` (enum, required) — The kind of the asset. - Allowed values: `compute`, `environment`, `accessKey`, `dockerRegistry`, `password`, `genericSecret`, `registry`, `s3`, `git`, `nfs`, `pvc`, `hostPath`, `workload-template`, `model`, `config-map`, `secret-volume`, `data-volume`, `ngcApiKey` - `createdBy` (string, required) — The user who created the asset. - `createdAt` (datetime, required) — The time at which the asset were created - `updatedBy` (string, required) — The user who updated the asset. - `updatedAt` (datetime, required) — The time at which the asset has been updated - `description` (string, optional) — The description of the asset. - `clusterId` (string, optional) — The id of the cluster. - `departmentId` (string, optional, nullable) — The id of the department. Must be specified for department scoped assets. - `projectId` (integer, optional, nullable) — The id of the project. Must be specified for project scoped assets. - `autoDelete` (boolean, optional, nullable, default: false) — The asset will be deleted automatically. This is intended for internal use. - `workloadSupportedTypes` (WorkloadSupportedTypes, optional) - `private` (boolean, optional, nullable, default: false) — The asset will be private to the user who created it. supported only for docker registry and generic secrets - `tenantId` (integer, optional, nullable) — The id of the tenant. - `deletedAt` (datetime, optional) — Deletion time of the asset (relevant for assets that sync to the cluster) - `deletedBy` (string, optional) — The user who deleted the asset - `projectName` (string, optional, nullable) — The name of the project that the asset is associated with, for project scoped assets. - `updateCount` (long, optional) — the value of an update counter of a data in the system. ### S3AssetSpec - `bucket` (string, optional, nullable) — The name of the bucket. (mandatory) - `path` (string, optional, nullable) — Local path within the workload to which the S3 bucket will be mapped. (mandatory) - `url` (string, optional, nullable) — The url of the S3 service provider. The default is the URL of the Amazon AWS S3 service. - `accessKeyAssetId` (string, optional, nullable) — ID of credentials asset of type access-key, for private S3 buckets. ### AssetUsageInfo Details about resources that use the asset. - `workspaces` (list of WorkloadRefAndStatus, optional) — workspaces that rely on this asset. - `trainings` (list of WorkloadRefAndStatus, optional) — trainings that rely on this asset. - `distributed` (list of WorkloadRefAndStatus, optional) — distributed trainings that rely on this asset. - `inferences` (list of WorkloadRefAndStatus, optional) — inferences that rely on this asset. - `templates` (list of AssetRef, optional) — templates that rely on this asset. - `assets` (AssetsUsageRef, optional) — list of other assets that use this asset. specifically, accessKeys used by s3, and passwords used by git. ### UsageTimesInfo Details about times the asset has been used by workloads, etc. - `lastUsedByWorkload` (datetime, optional, nullable) — The time at which the asset were last used by a workload of given type and project. This field requires complyToProject and complyToWorkload types flags of the API, to specify the workload type and project id. ### ComplianceInfo Compliance of an asset with creation of workloads in a given project. - `imposed` (boolean, optional) — Whether the asset is imposed on created workloads by the administrator. - `compliance` (boolean, optional) — Whether the asset info complies with creation of workloads in the project. - `reason` (list of ComplianceInfoReason, optional) — For asset that does not comply, specify one or more reason why. ### AssetClusterStatusInfo - `status` (enum, optional) - Allowed values: `Creating`, `Deleting`, `Failed`, `Deletion failed`, `Issues found`, `No issues found`, `No status`, `Updating` - `issues` (list of AssetClusterStatusIssue, optional) - `message` (string, optional) - `url` (string, optional) ### WorkloadSupportedTypes - `inference` (boolean, optional, nullable) — Is inference a supported workload type. - `workspace` (boolean, optional, nullable) — Is workspace a supported workload type. - `training` (boolean, optional, nullable) — Is training a supported workload type. - `distributed` (boolean, optional, nullable) — Is distributed a supported workload type. - `distFramework` (enum, optional, nullable) — The distributed training framework used in the workload. Note: The TF, XGBoost and JAX frameworks are deprecated and will be removed in a future release. - Allowed values: `MPI`, `PyTorch`, `TF`, `XGBoost`, `JAX` ### WorkloadRefAndStatus Id, name and status of a workspace or training - `id` (string, required) — A unique ID of the workload. - `name` (string, required) — The name of the workload. - `status` (string, optional) — The status of the workspace or training. ### AssetRef Reference information about an asset. - `id` (string, required) — Unique identifier of the asset. - `name` (string, required) — The name of the asset. ### AssetsUsageRef Reference information about usage of assets by other assets, for example datasources that uses credentials. - `environment` (EnvironmentAssetRef, optional, nullable) — environment asset. - `environments` (list of AssetRef, optional) - `compute` (AssetsUsageRefCompute, optional) - `computes` (list of AssetRef, optional) - `datasources` (list of AssetDatasourceRef, optional) ### ComplianceInfoReason An issue preventing the use of a template or asset for creating workloads in a given project. - `details` (string, required) — Explanation of the issue. - `field` (string, optional, nullable) — For policy related issues, includes a json path to the field in the policy for which the issue applies. - `rule` (enum, optional) — Indicates which validation rule (e.g., min, max, step, options, required, canEdit, canAdd) conflicted with policy restrictions, causing the asset or template to be rejected. - Allowed values: `min`, `max`, `step`, `options`, `required`, `canEdit`, `canAdd`, `locked`, `blocked` ### AssetClusterStatusIssue - `scopeId` (string, required) — The id of the cluster, department or project, depending on the scope type. - `scopeType` (enum, required) - Allowed values: `system`, `tenant`, `cluster`, `department`, `project` - `issue` (enum, required) - Allowed values: `ReplicationError`, `LostPhase`, `NotCreated` ### EnvironmentAssetRef Reference information about environment asset. - `id` (string, required) — Unique identifier of the asset. - `name` (string, required) — The name of the asset. - `toolTypes` (list of enum, optional, nullable) — the type of tools as defined in the connections of the environment asset. - Allowed values: `jupyter-notebook`, `pycharm`, `visual-studio-code`, `tensorboard`, `rstudio`, `mlflow`, `custom`, `wandb`, `matlab`, `comet`, `chatbot-ui` ### AssetsUsageRefCompute - `id` (string, required) — Unique identifier of the asset. - `name` (string, required) — The name of the asset. ### AssetDatasourceRef Reference information about a datasource asset. - `id` (string, required) — Unique identifier of the asset. - `name` (string, required) — The name of the asset. - `kind` (enum, required) — The kind of the asset. - Allowed values: `compute`, `environment`, `accessKey`, `dockerRegistry`, `password`, `genericSecret`, `registry`, `s3`, `git`, `nfs`, `pvc`, `hostPath`, `workload-template`, `model`, `config-map`, `secret-volume`, `data-volume`, `ngcApiKey` - `overrides` (DataSourceOverrides, optional) — specific fields to override in the data source. ### DataSourceOverrides specific fields to override in the data source. - `containerPath` (string, optional, nullable) ## Examples **Response** ```json { "entries": [ { "meta": { "name": "my-asset", "scope": "tenant", "id": "a418ed33-9399-48c0-a890-122cadd13bfd", "kind": "s3", "createdBy": "test@run.ai", "createdAt": "2023-02-23T14:25:36.707685Z", "updatedBy": "test@run.ai", "updatedAt": "2023-02-23T14:25:36.707685Z", "workloadSupportedTypes": { "workspace": false, "training": false, "distributed": true, "distFramework": "TF" } }, "spec": { "bucket": "my-bucket", "path": "/container/my-bucket", "url": "https://s3.amazonaws.com", "accessKeyAssetId": "string" }, "usedBy": { "workspaces": [ { "id": "string", "name": "my-workload-name", "status": "string" } ], "trainings": [ { "id": "string", "name": "my-workload-name", "status": "string" } ], "distributed": [ { "id": "string", "name": "my-workload-name", "status": "string" } ], "inferences": [ { "id": "string", "name": "my-workload-name", "status": "string" } ], "templates": [ { "id": "string", "name": "my-asset" } ], "assets": { "environment": { "id": "string", "name": "my-asset", "toolTypes": [ "jupyter-notebook" ] }, "environments": [ { "id": "string", "name": "my-asset" } ], "compute": { "id": "string", "name": "my-asset" }, "computes": [ { "id": "string", "name": "my-asset" } ], "datasources": [ { "id": "string", "name": "my-asset", "kind": "compute", "overrides": { "containerPath": "/container/directory" } } ] } }, "usageTimes": { "lastUsedByWorkload": "2024-01-15T09:30:00Z" }, "compliance": { "imposed": true, "compliance": true, "reason": [ { "details": "string", "field": "string", "rule": "min" } ] }, "status": { "status": "Creating", "issues": [ { "scopeId": "string", "scopeType": "system", "issue": "ReplicationError" } ], "message": "string", "url": "string" } } ] } ``` **SDK Code** ```python import requests url = "http://localhost:8080/api/v1/asset/datasource/s3" headers = {"Authorization": "Bearer "} response = requests.get(url, headers=headers) print(response.json()) ``` ```javascript const url = 'http://localhost:8080/api/v1/asset/datasource/s3'; const options = {method: 'GET', headers: {Authorization: 'Bearer '}}; try { const response = await fetch(url, options); const data = await response.json(); console.log(data); } catch (error) { console.error(error); } ``` ```go package main import ( "fmt" "net/http" "io" ) func main() { url := "http://localhost:8080/api/v1/asset/datasource/s3" req, _ := http.NewRequest("GET", url, nil) req.Header.Add("Authorization", "Bearer ") res, _ := http.DefaultClient.Do(req) defer res.Body.Close() body, _ := io.ReadAll(res.Body) fmt.Println(res) fmt.Println(string(body)) } ``` ```ruby require 'uri' require 'net/http' url = URI("http://localhost:8080/api/v1/asset/datasource/s3") http = Net::HTTP.new(url.host, url.port) request = Net::HTTP::Get.new(url) request["Authorization"] = 'Bearer ' response = http.request(request) puts response.read_body ``` ```java import com.mashape.unirest.http.HttpResponse; import com.mashape.unirest.http.Unirest; HttpResponse response = Unirest.get("http://localhost:8080/api/v1/asset/datasource/s3") .header("Authorization", "Bearer ") .asString(); ``` ```php request('GET', 'http://localhost:8080/api/v1/asset/datasource/s3', [ 'headers' => [ 'Authorization' => 'Bearer ', ], ]); echo $response->getBody(); ``` ```csharp using RestSharp; var client = new RestClient("http://localhost:8080/api/v1/asset/datasource/s3"); var request = new RestRequest(Method.GET); request.AddHeader("Authorization", "Bearer "); IRestResponse response = client.Execute(request); ``` ```swift import Foundation let headers = ["Authorization": "Bearer "] let request = NSMutableURLRequest(url: NSURL(string: "http://localhost:8080/api/v1/asset/datasource/s3")! as URL, cachePolicy: .useProtocolCachePolicy, timeoutInterval: 10.0) request.httpMethod = "GET" request.allHTTPHeaderFields = headers let session = URLSession.shared let dataTask = session.dataTask(with: request as URLRequest, completionHandler: { (data, response, error) -> Void in if (error != nil) { print(error as Any) } else { let httpResponse = response as? HTTPURLResponse print(httpResponse) } }) dataTask.resume() ```