POST /mltk/start_container

Description. Start a container (Docker) or deployment (Kubernetes). The handler resolves the image from docker_images.conf, injects API token and observability env vars, applies HPA overrides for Kubernetes, and probes the resulting endpoint.

Body (JSON).

{
  "payload": {
    "image":   "golden-cpu",
    "model":   "my_model",
    "runtime": "nvidia",
    "cluster": "docker",
    "mode":    "DEV",

    "hpa_enabled":          "0",
    "min_cpu":              "250m",
    "max_cpu":              "1000m",
    "min_memory":           "512Mi",
    "max_memory":           "2Gi",
    "min_replicas":         "1",
    "max_replicas":         "3",
    "cpu_threshold_percent":"80"
  }
}

Response (200).

{
  "status":  "success",
  "message": "The container started successfully",
  "updated_config": {
    "id":               "container_abc123",
    "api_url":          "http://10.0.0.5:5000",
    "api_url_external": "http://host.example.com:5000",
    "jupyter_url":      "http://10.0.0.5:8888",
    "status":           "Waiting for Url to start",
    "cluster":          "docker",
    "mode":             "DEV"
  }
}

Example.

curl -sk -u "$SPLUNK_USER:$SPLUNK_PASSWORD" -X POST 
  -H "Content-Type: application/json" 
  --data @- "$SPLUNK_HOST/servicesNS/nobody/Splunk_ML_Toolkit/mltk/start_container?output_mode=json" <<'JSON'
{
  "payload": {
    "image":   "golden-cpu",
    "model":   "my_model",
    "cluster": "docker",
    "mode":    "DEV"
  }
}
JSON