Start a cluster

import requests

url = "https://platform.ankra.app/api/v1/clusters/{cluster_id}/provision"

headers = {"Authorization": "Bearer <token>"}

response = requests.post(url, headers=headers)

print(response.text)

const options = {method: 'POST', headers: {Authorization: 'Bearer <token>'}};

fetch('https://platform.ankra.app/api/v1/clusters/{cluster_id}/provision', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

{
  "marked_to_start_at": "2023-11-07T05:31:56Z",
  "created_operations": 0,
  "scope": "all",
  "state_restore": "requested",
  "state_snapshot_id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
  "state_snapshot": {
    "captured_at": "2023-11-07T05:31:56Z",
    "id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
    "trigger": "<string>"
  },
  "captured_state_snapshots": 0,
  "state_restore_message": "<string>"
}

{
  "detail": [
    {
      "loc": [
        "<string>"
      ],
      "msg": "<string>",
      "type": "<string>"
    }
  ]
}

Clusters

Start a cluster

Starts a stopped cluster, retries a start whose control plane never came up (the cluster offline or provider_without_access), or brings back parked workers with the default scope; any other state answers 409. Pass scope=control_plane to start only the control plane (the default starts everything), and restore_state=false to boot an empty etcd instead of restoring a captured cluster state snapshot. By default a cluster whose stop deleted its VMs restores a restore already requested, otherwise the newest snapshot captured after its last start (the stop’s own capture, otherwise the newest scheduled backup); an older one is restored only when named with restore_snapshot_id. A paused cluster keeps its etcd and gets no new restore, unless restore_over_kept_etcd=true is passed with restore_snapshot_id to confirm that the kept control-plane machines, and their etcd, are lost. The answer’s state_restore, state_snapshot and state_restore_message say which snapshot is restored. Answers 409 when the cluster cannot be started in its current state.

POST

api

clusters

{cluster_id}

provision

Start a cluster

import requests

url = "https://platform.ankra.app/api/v1/clusters/{cluster_id}/provision"

headers = {"Authorization": "Bearer <token>"}

response = requests.post(url, headers=headers)

print(response.text)

const options = {method: 'POST', headers: {Authorization: 'Bearer <token>'}};

fetch('https://platform.ankra.app/api/v1/clusters/{cluster_id}/provision', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

{
  "marked_to_start_at": "2023-11-07T05:31:56Z",
  "created_operations": 0,
  "scope": "all",
  "state_restore": "requested",
  "state_snapshot_id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
  "state_snapshot": {
    "captured_at": "2023-11-07T05:31:56Z",
    "id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
    "trigger": "<string>"
  },
  "captured_state_snapshots": 0,
  "state_restore_message": "<string>"
}

{
  "detail": [
    {
      "loc": [
        "<string>"
      ],
      "msg": "<string>",
      "type": "<string>"
    }
  ]
}

Authorizations

Authorization

string

header

required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Headers

authorization

string | null

x-ankra-organisation-id

string | null

Path Parameters

cluster_id

string<uuid4>

required

Query Parameters

scope

string | null

'control_plane' to only provision control plane, or 'all' (default)

Response

Successful Response

marked_to_start_at

string<date-time>

required

created_operations

integer

default:0

scope

string

default:all

state_restore

enum<string>

What the first control plane does with the cluster's etcd: requested (it restores state_snapshot), none (nothing captured, or the cluster kept its etcd), skipped (restore_state=false: it bootstraps an empty etcd).

Available options:

requested,

none,

skipped

state_snapshot_id

string<uuid>

The restored snapshot's id, set with state_restore=requested.

state_snapshot

State Snapshot · object

The restored snapshot, set with state_restore=requested.

Show child attributes

captured_state_snapshots

integer

default:0

How many captured state snapshots of the cluster a start could restore, the restored one included.

state_restore_message

string

One sentence on what happens to the cluster's state: the snapshot restored (id, trigger, capture time), or that the cluster boots with an empty etcd and how many captured snapshots are left unrestored.

API reference

Endpoints

Start a cluster

Authorizations

Headers

Path Parameters

Query Parameters

Response