Scale service

Set the service’s instance count and, optionally, its autoscaling targets and ceiling.

PATCH /api/services/{id}/scaling HTTPS

Request & response examples

Request example Server-side

curl --fail-with-body --request PATCH \
  'https://app.helicarrier.xyz/api/services/YOUR_ID/scaling' \
  -H "Authorization: Bearer $HELI_API_KEY" \
  -H 'Content-Type: application/json' \
  --data '{
  "instances": 2,
  "maxInstances": 4,
  "autoscaleEnabled": true,
  "autoscaleCpuPct": 70
}'

const response = await fetch(
  "https://app.helicarrier.xyz/api/services/YOUR_ID/scaling",
  {
    method: "PATCH",
    headers: {
      Authorization: `Bearer ${process.env.HELI_API_KEY}`,
      "Content-Type": "application/json"
    },
    body: JSON.stringify({
      "instances": 2,
      "maxInstances": 4,
      "autoscaleEnabled": true,
      "autoscaleCpuPct": 70
    })
  }
);
const data = await response.json();
if (!response.ok) {
  throw new Error(`HTTP ${response.status}: ${data.error}`);
}
// Inspect data; avoid logging secret responses.

import os
import requests

response = requests.request(
    "PATCH",
    "https://app.helicarrier.xyz/api/services/YOUR_ID/scaling",
    headers={
        "Authorization": f"Bearer {os.environ['HELI_API_KEY']}"
    },
    json={
      "instances": 2,
      "maxInstances": 4,
      "autoscaleEnabled": True,
      "autoscaleCpuPct": 70
    },
    timeout=30,
)
response.raise_for_status()
data = response.json()
# Inspect data; avoid logging secret responses.

Response example 200

{
  "scaling": {
    "minReplicas": 1,
    "maxReplicas": 4,
    "desiredReplicas": 2,
    "instances": 2,
    "autoscaleEnabled": true,
    "autoscaleCpuPct": 70,
    "autoscaleMemPct": 0,
    "autoscalable": true
  },
  "redeploying": true
}

Illustrative values · selected fields

Response details ↓ OpenAPI ↗

Authorization

An account key with access, or a project/service key with settings:write. The resource must be within the key’s project, service, and environment scope.

Authorization: Bearer <key> Key setup ↗

Path parameters

id string required

Service id.

Request body application/json

instances integer optional

Instances to run (minimum 1). With autoscaling on this is the floor.

maxInstances integer optional

Ceiling autoscaling may grow to. Must be >= instances.

autoscaleEnabled boolean optional

Turn autoscaling on or off.

autoscaleCpuPct integer optional

Target CPU utilisation percent per instance (0 = ignore CPU).

autoscaleMemPct integer optional

Target memory utilisation percent per instance (0 = ignore memory).

Response 200

Returns the saved scaling configuration and whether an immediate redeployment was applied. The response uses minReplicas and maxReplicas for the requested floor and ceiling.

Examples show selected response fields with illustrative values. Your response can contain additional fields.

scaling object

Saved scaling configuration and effective capacity.

Child fields

minReplicas integer
Configured instance floor.
maxReplicas integer
Configured autoscaling ceiling.
desiredReplicas integer
Desired instance count selected by the scaler.
instances integer
Effective number of instances.
autoscaleEnabled boolean
Whether autoscaling is enabled.
autoscaleCpuPct integer
CPU utilization target; zero disables this signal.
autoscaleMemPct integer
Memory utilization target; zero disables this signal.
autoscalable boolean
Whether this service supports autoscaling.

redeploying boolean

Whether the operation applied or queued a redeployment.

Errors

400

Invalid parameters. Check the required fields, types, and resource configuration.

401

The API key is missing, invalid, or revoked.

403

The caller or key scope does not permit this operation.

500

The operation could not be completed. Check resource state before retrying a write.

Error handling and safe retries

ALSO AVAILABLE VIA MCP

`scale_service`

Required MCP arguments: id. Send path, query, and body fields together as tool arguments.

Connect your agent ↗