Pod

Management of PODs is also available via API, however, the creation of a POD is not available via API nor via the panel, the ServerlessGPU service does not intend to represent a panel for managing the Kubernetes cluster.

If you are looking for a management panel for your Kubernetes cluster, the Seeweb Team strongly recommends solutions similar to Rancher.

View Your PODs

You can view all your PODs, to do this, the following request is available:

HTTP Request

GET /k8sgpu/v1/pod/

HTTP Response

[
  {
    "uuid": "ad5eabd2-396d-44c2-869e-c82a4df87f18",
    "remote_reference_name": "inference-service-5cfb7489f4-2d4xj",
    "remote_reference_namespace": "remote-gpu",
    "local_reference_name": "69d416b2-1c02-4910-9fcb-ae0d0da408f0",
    "local_reference_namespace": "skg00018-gpu",
    "status": "terminated",
    "online": false,
    "active": true,
    "succeeded": false,
    "failed": false,
    "creation_date": "2024-08-12T14:48:00+00:00",
    "deactivated_from": null,
    "runtime_class": "seeweb-nvidia-1xa6000",
    "user": "skgu00004",
    "cluster": "skg00018",
    "in_deletion": false
  },
  {
    "uuid": "18009a59-b1c8-4ab9-8ad5-a0c7c2efe6eb",
    "remote_reference_name": "inference-service-5cfb7489f4-5xvs9",
    "remote_reference_namespace": "remote-gpu",
    "local_reference_name": "d7b1c84f-9441-43bd-820c-7a06ed2db094",
    "local_reference_namespace": "skg00018-gpu",
    "status": "terminated",
    "online": false,
    "active": true,
    "succeeded": false,
    "failed": false,
    "creation_date": "2024-08-22T12:37:44+00:00",
    "deactivated_from": null,
    "runtime_class": "seeweb-nvidia-1xa6000",
    "user": "skgu00004",
    "cluster": "skg00018",
    "in_deletion": false
  },
]

Filter Your PODs by Tenant

You can filter your PODs by Tenant using the following parameter in the query:

GET /k8sgpu/v1/pod/?cluster=skg00018

View a Specific POD

Of course, you can also view a specific POD:

HTTP Request

GET /k8sgpu/v1/pod/{pod uuid}

HTTP Response

{
  "uuid": "ad5eabd2-396d-44c2-869e-c82a4df87f18",
  "remote_reference_name": "inference-service-5cfb7489f4-2d4xj",
  "remote_reference_namespace": "remote-gpu",
  "local_reference_name": "69d416b2-1c02-4910-9fcb-ae0d0da408f0",
  "local_reference_namespace": "skg00018-gpu",
  "status": "terminated",
  "online": false,
  "active": true,
  "succeeded": false,
  "failed": false,
  "creation_date": "2024-08-12T14:48:00+00:00",
  "deactivated_from": null,
  "runtime_class": "seeweb-nvidia-1xa6000",
  "user": "skgu00004",
  "cluster": "skg00018",
  "in_deletion": false
}

Delete a POD

As in the panel, it is also possible to request the deletion of a POD via API, to do this you need to send the following request:

HTTP Request:

DELETE /k8sgpu/v1/pod/{pod uuid}