Create a CoreCluster
Creates a CoreCluster in a region on a package, with up to ten first sites, charged at the package's price from the moment it is created. Follow it with GET / until it has finished. Only a create that is accepted counts towards the rate limit, which your whole company shares.
Body
The server type. optimised is accepted only while it is offered to your company. Defaults to general.
generaloptimisedThe CoreCluster's name: letters, digits, spaces and . & ' ( ) -, starting with a letter or a digit. No other CoreCluster of your company may have it. Up to 100 characters.
The key of a CoreCluster package, such as xsmall. GET /corestack/packages lists them.
A region that is open for the package. GET /corestack/regions lists them.
ukusnlausThe first sites to create with the CoreCluster. Add the rest once it is ready. Up to 10 items.
Fields of each item
empty for a site with no application installed, fresh for a fresh install.
emptyfreshThe site's public domain, without www.. Each domain may appear once in the request. Up to 253 characters.
The settings of a fresh install. Read only when content is fresh.
A MariaDB version the platform offers. Left out, the platform's default.
["eol"] to accept a Node.js version past its end of life, which is refused ack_required without it. Up to 1 items.
eolThe Node.js major version for the site, as a number such as 22, from the ones offered for the platform. Left out, the default. Where a new site cannot choose one yet, it is refused runtime_not_adopted. 1 to 99.
A PHP version the platform offers, such as 8.3. Left out, the platform's default.
A platform the package offers, such as wordpress, laravel or magento2.
The size of the site, such as S, M or L. It must be one the platform has and the package offers.
Magento 2 only. Whether the site starts with Varnish. Left out, the platform decides.
Fields
A Varnish version the platform offers, such as 7.7. Only with enabled true.
Returns
201 The CoreCluster is being created. sites lists each first site with its own operation.
202 The create was accepted and its first step is still being confirmed. The body is the same as for 201, with the operation's status pending. Poll the operation until it finishes.
Fields of the 201 answer
A CoreCluster. Every answer that carries a CoreCluster carries this object. Other fields may be present; the ones named in required are always sent.
Fields
What can be done with the CoreCluster now.
Fields
Whether something can be done now and, when it cannot, why not.
Fields
May add grows, true when the CoreCluster is full and would grow first under automatic upgrades.
Fields
Whether something can be done now and, when it cannot, why not.
Fields
Adds downgrade, true when a smaller package can be chosen.
Fields
What the package allows.
Fields
Sent only by GET /corestack/clusters/{cluster}, while billing is on for your company.
Fields
The billing account the CoreCluster is charged to. Null when nothing charges it yet, or you cannot see the account. At least 1.
How much of the package the CoreCluster's sites use.
Fields
What the usage figures are: measured (p95), estimated from the sizes of the sites (estimate), or both (mixed).
p95mixedestimateOne meter. The figures are in the unit its field names say; percent is what is used of the capacity.
Fields
ok under 75 per cent, warning from 75 and blocked from 90.
okwarningblockedFormat: date-time.
The worst level of the meters.
okwarningblockedOne meter. The figures are in the unit its field names say; percent is what is used of the capacity.
Fields
ok under 75 per cent, warning from 75 and blocked from 90.
okwarningblockedThe package the figures are measured against. Null for a CoreCluster with no package.
Fields
Fields
One meter. The figures are in the unit its field names say; percent is what is used of the capacity.
Fields
ok under 75 per cent, warning from 75 and blocked from 90.
okwarningblockedFormat: date-time.
How many environments besides production the CoreCluster has.
The {cluster} of every /corestack/clusters/{cluster} endpoint. At least 1.
customer for a CoreCluster of your own.
customerplatformFields
arm64amd64Where the CoreCluster runs.
How much log storage the CoreCluster uses. Null when its logs are not collected.
The change running on the CoreCluster now, or the one that left it failed. Null otherwise.
The CoreCluster's package. Null when it has none, and while packages are off for your company.
Fields
The changes waiting for the next maintenance window of the CoreCluster's sites.
Fields
Format: date-time.
Format: date-time.
Fields
The region key, as a create takes it.
What the sizes of the CoreCluster's sites add up to.
Fields
How many sites the CoreCluster has, and how they are doing.
Fields
Environments besides production.
Sites with a domain attached.
Sites with no domain attached yet.
Made from the name when the CoreCluster is created. It never changes.
The short form of the CoreCluster's snapshot settings and storage.
Fields
The address the CoreCluster's sites are reached on over SSH.
creatingreadydegradedmigratingdeletingfailedWhy the CoreCluster is degraded, migrating or failed, in plain words.
When the list of sites, the usage and the live status were last read.
Fields
Format: date-time.
Format: date-time.
Format: date-time.
What the sites use now. A figure is null where nothing was measured.
Fields
Format: date-time.
A change the Control Panel is making or has made. A save or an action that starts work answers one in operation; read it again by id with GET / until status is completed or failed. Both answers have this shape; the few endpoints that answer a different one say so.
Fields
The cluster the operation runs on.
Why the operation failed, in plain words. Null unless status is failed.
A stable code for the failure.
True when the change applies without a maintenance window.
What the operation did differently on purpose. Empty when nothing.
Fields of each item
Who asked for the change.
Present on some site creations and copies.
Always null here. A one-time result is read once with the reveal endpoint.
nonependingavailableviewedWhat a retry needs first, if anything.
Who revealed a one-time result and when, when there was one.
siteclusterbatchThe website the operation belongs to. Null for a cluster's own operation.
Where a change is. completed and failed are final; every other status means the change is still running and blocks another change to the same site.
pendingcommittedrolling_outexecutingcompletedfailedThe key of the step the operation is at. Null once it has completed.
Fields of each item
One of done, current, todo, failed or skipped.
The kind of change, such as crons_change or redirects_change.
Each first site with the operation that creates it. Empty when the request listed none.
Fields of each item
A change the Control Panel is making or has made. A save or an action that starts work answers one in operation; read it again by id with GET / until status is completed or failed. Both answers have this shape; the few endpoints that answer a different one say so.
Fields
The cluster the operation runs on.
Why the operation failed, in plain words. Null unless status is failed.
A stable code for the failure.
True when the change applies without a maintenance window.
What the operation did differently on purpose. Empty when nothing.
Fields of each item
Who asked for the change.
Present on some site creations and copies.
Always null here. A one-time result is read once with the reveal endpoint.
nonependingavailableviewedWhat a retry needs first, if anything.
Who revealed a one-time result and when, when there was one.
siteclusterbatchThe website the operation belongs to. Null for a cluster's own operation.
Where a change is. completed and failed are final; every other status means the change is still running and blocks another change to the same site.
pendingcommittedrolling_outexecutingcompletedfailedThe key of the step the operation is at. Null once it has completed.
Fields of each item
One of done, current, todo, failed or skipped.
The kind of change, such as crons_change or redirects_change.
One site or environment in full. The summaries (domains_summary, access_summary and the others) are short forms of what the endpoints of each feature answer in full. A site or environment row as lists and progress answers show it. Other fields may be present; the ones named here are always sent.
Fields
healthydegradeddownapplyingunknownnot_liveFields
customer for a site in one of your CoreClusters, platform for a standalone CoreStack site.
The site row id, the {site} of every /corestack/sites/{site} endpoint.
On a read (a site, the list of sites, a CoreCluster's sites, the overview): corecluster for a site in one of your CoreClusters, corestack for a standalone CoreStack site. The answer to a change that carries the row, such as creating a site, can send corestack for either; hosting.cluster_kind tells the two apart on every answer.
coreclustercorestackFields
Fields
production once a domain is attached, staging before.
productionstagingunknownThe change running on the row now, or null.
The package of a standalone CoreStack site, with its key, code and name. Null for a CoreCluster site.
The changes waiting for the row's next maintenance window.
Fields
Fields
Fields
The row's own address on the platform.
The restore running on the row now, or null.
The site the row belongs to. For a production row, the row itself.
The row's platform identifier.
A summary of the row's snapshots.
Fields
The size of the row, such as S, M or L.
The size of a staging environment. Null for a production row.
Fields
What you can do with the row now. Each entry says whether it is available and, when not, why.
A cluster by its id and name. A standalone CoreStack site reads CoreStack as the name.
Fields
How many scheduled tasks and redirects the row has, each while its feature is on.
How to connect to the database. The password is never in this answer.
When the site was deleted and until when it can be restored. Null for a site that is not deleted.
Fields of each item
The environment's platform identifier.
The environment's own address on the platform.
The site row id, the {site} every /corestack/sites/{site} endpoint takes.
The environment's role, such as production or staging.
Where the row is in its life, such as active or deleted.
The status last read from the cluster. Null until one has been read.
The row's most recent operations.
Fields of each item
The cluster the operation runs on.
Why the operation failed, in plain words. Null unless status is failed.
A stable code for the failure.
True when the change applies without a maintenance window.
What the operation did differently on purpose. Empty when nothing.
Fields of each item
Who asked for the change.
Present on some site creations and copies.
Always null here. A one-time result is read once with the reveal endpoint.
nonependingavailableviewedWhat a retry needs first, if anything.
Who revealed a one-time result and when, when there was one.
siteclusterbatchThe website the operation belongs to. Null for a cluster's own operation.
Where a change is. completed and failed are final; every other status means the change is still running and blocks another change to the same site.
pendingcommittedrolling_outexecutingcompletedfailedThe key of the step the operation is at. Null once it has completed.
Fields of each item
One of done, current, todo, failed or skipped.
The kind of change, such as crons_change or redirects_change.
The row's platform address, and whether HTTP authentication protects it.
The PHP, MariaDB and Node.js versions with the versions the row can change to.
The environment's own switches, as PATCH / sets them.
Fields
Whether the environment deploys automatically from Git.
Whether HTTP authentication is on for the whole environment.
Whether scheduled snapshots are on.
How to connect over SSH and SFTP.
Fields of the 202 answer
A CoreCluster. Every answer that carries a CoreCluster carries this object. Other fields may be present; the ones named in required are always sent.
Fields
What can be done with the CoreCluster now.
Fields
Whether something can be done now and, when it cannot, why not.
Fields
May add grows, true when the CoreCluster is full and would grow first under automatic upgrades.
Fields
Whether something can be done now and, when it cannot, why not.
Fields
Adds downgrade, true when a smaller package can be chosen.
Fields
What the package allows.
Fields
Sent only by GET /corestack/clusters/{cluster}, while billing is on for your company.
Fields
The billing account the CoreCluster is charged to. Null when nothing charges it yet, or you cannot see the account. At least 1.
How much of the package the CoreCluster's sites use.
Fields
What the usage figures are: measured (p95), estimated from the sizes of the sites (estimate), or both (mixed).
p95mixedestimateOne meter. The figures are in the unit its field names say; percent is what is used of the capacity.
Fields
ok under 75 per cent, warning from 75 and blocked from 90.
okwarningblockedFormat: date-time.
The worst level of the meters.
okwarningblockedOne meter. The figures are in the unit its field names say; percent is what is used of the capacity.
Fields
ok under 75 per cent, warning from 75 and blocked from 90.
okwarningblockedThe package the figures are measured against. Null for a CoreCluster with no package.
Fields
Fields
One meter. The figures are in the unit its field names say; percent is what is used of the capacity.
Fields
ok under 75 per cent, warning from 75 and blocked from 90.
okwarningblockedFormat: date-time.
How many environments besides production the CoreCluster has.
The {cluster} of every /corestack/clusters/{cluster} endpoint. At least 1.
customer for a CoreCluster of your own.
customerplatformFields
arm64amd64Where the CoreCluster runs.
How much log storage the CoreCluster uses. Null when its logs are not collected.
The change running on the CoreCluster now, or the one that left it failed. Null otherwise.
The CoreCluster's package. Null when it has none, and while packages are off for your company.
Fields
The changes waiting for the next maintenance window of the CoreCluster's sites.
Fields
Format: date-time.
Format: date-time.
Fields
The region key, as a create takes it.
What the sizes of the CoreCluster's sites add up to.
Fields
How many sites the CoreCluster has, and how they are doing.
Fields
Environments besides production.
Sites with a domain attached.
Sites with no domain attached yet.
Made from the name when the CoreCluster is created. It never changes.
The short form of the CoreCluster's snapshot settings and storage.
Fields
The address the CoreCluster's sites are reached on over SSH.
creatingreadydegradedmigratingdeletingfailedWhy the CoreCluster is degraded, migrating or failed, in plain words.
When the list of sites, the usage and the live status were last read.
Fields
Format: date-time.
Format: date-time.
Format: date-time.
What the sites use now. A figure is null where nothing was measured.
Fields
Format: date-time.
A change the Control Panel is making or has made. A save or an action that starts work answers one in operation; read it again by id with GET / until status is completed or failed. Both answers have this shape; the few endpoints that answer a different one say so.
Fields
The cluster the operation runs on.
Why the operation failed, in plain words. Null unless status is failed.
A stable code for the failure.
True when the change applies without a maintenance window.
What the operation did differently on purpose. Empty when nothing.
Fields of each item
Who asked for the change.
Present on some site creations and copies.
Always null here. A one-time result is read once with the reveal endpoint.
nonependingavailableviewedWhat a retry needs first, if anything.
Who revealed a one-time result and when, when there was one.
siteclusterbatchThe website the operation belongs to. Null for a cluster's own operation.
Where a change is. completed and failed are final; every other status means the change is still running and blocks another change to the same site.
pendingcommittedrolling_outexecutingcompletedfailedThe key of the step the operation is at. Null once it has completed.
Fields of each item
One of done, current, todo, failed or skipped.
The kind of change, such as crons_change or redirects_change.
Each first site with the operation that creates it. Empty when the request listed none.
Fields of each item
A change the Control Panel is making or has made. A save or an action that starts work answers one in operation; read it again by id with GET / until status is completed or failed. Both answers have this shape; the few endpoints that answer a different one say so.
Fields
The cluster the operation runs on.
Why the operation failed, in plain words. Null unless status is failed.
A stable code for the failure.
True when the change applies without a maintenance window.
What the operation did differently on purpose. Empty when nothing.
Fields of each item
Who asked for the change.
Present on some site creations and copies.
Always null here. A one-time result is read once with the reveal endpoint.
nonependingavailableviewedWhat a retry needs first, if anything.
Who revealed a one-time result and when, when there was one.
siteclusterbatchThe website the operation belongs to. Null for a cluster's own operation.
Where a change is. completed and failed are final; every other status means the change is still running and blocks another change to the same site.
pendingcommittedrolling_outexecutingcompletedfailedThe key of the step the operation is at. Null once it has completed.
Fields of each item
One of done, current, todo, failed or skipped.
The kind of change, such as crons_change or redirects_change.
One site or environment in full. The summaries (domains_summary, access_summary and the others) are short forms of what the endpoints of each feature answer in full. A site or environment row as lists and progress answers show it. Other fields may be present; the ones named here are always sent.
Fields
healthydegradeddownapplyingunknownnot_liveFields
customer for a site in one of your CoreClusters, platform for a standalone CoreStack site.
The site row id, the {site} of every /corestack/sites/{site} endpoint.
On a read (a site, the list of sites, a CoreCluster's sites, the overview): corecluster for a site in one of your CoreClusters, corestack for a standalone CoreStack site. The answer to a change that carries the row, such as creating a site, can send corestack for either; hosting.cluster_kind tells the two apart on every answer.
coreclustercorestackFields
Fields
production once a domain is attached, staging before.
productionstagingunknownThe change running on the row now, or null.
The package of a standalone CoreStack site, with its key, code and name. Null for a CoreCluster site.
The changes waiting for the row's next maintenance window.
Fields
Fields
Fields
The row's own address on the platform.
The restore running on the row now, or null.
The site the row belongs to. For a production row, the row itself.
The row's platform identifier.
A summary of the row's snapshots.
Fields
The size of the row, such as S, M or L.
The size of a staging environment. Null for a production row.
Fields
What you can do with the row now. Each entry says whether it is available and, when not, why.
A cluster by its id and name. A standalone CoreStack site reads CoreStack as the name.
Fields
How many scheduled tasks and redirects the row has, each while its feature is on.
How to connect to the database. The password is never in this answer.
When the site was deleted and until when it can be restored. Null for a site that is not deleted.
Fields of each item
The environment's platform identifier.
The environment's own address on the platform.
The site row id, the {site} every /corestack/sites/{site} endpoint takes.
The environment's role, such as production or staging.
Where the row is in its life, such as active or deleted.
The status last read from the cluster. Null until one has been read.
The row's most recent operations.
Fields of each item
The cluster the operation runs on.
Why the operation failed, in plain words. Null unless status is failed.
A stable code for the failure.
True when the change applies without a maintenance window.
What the operation did differently on purpose. Empty when nothing.
Fields of each item
Who asked for the change.
Present on some site creations and copies.
Always null here. A one-time result is read once with the reveal endpoint.
nonependingavailableviewedWhat a retry needs first, if anything.
Who revealed a one-time result and when, when there was one.
siteclusterbatchThe website the operation belongs to. Null for a cluster's own operation.
Where a change is. completed and failed are final; every other status means the change is still running and blocks another change to the same site.
pendingcommittedrolling_outexecutingcompletedfailedThe key of the step the operation is at. Null once it has completed.
Fields of each item
One of done, current, todo, failed or skipped.
The kind of change, such as crons_change or redirects_change.
The row's platform address, and whether HTTP authentication protects it.
The PHP, MariaDB and Node.js versions with the versions the row can change to.
The environment's own switches, as PATCH / sets them.
Fields
Whether the environment deploys automatically from Git.
Whether HTTP authentication is on for the whole environment.
Whether scheduled snapshots are on.
How to connect over SSH and SFTP.
If it is refused
| Status | Reason | What to do |
|---|---|---|
401 | none | Send a valid token as Authorization: Bearer <token>. |
403 | company_unavailable | Your company is unavailable. |
403 | scope_required | Use a token with the scope named in scope. |
403 | two_factor_required | Turn on two-factor on your account, then send again. |
404 | feature_off | This feature is not on for your organisation yet. |
409 | cluster_busy | Another change to the cluster is running. Wait for it to finish, then try again. |
409 | compute_blocked | There is no server capacity for this size at the moment. Choose another size, or contact support. |
409 | duplicate | The same thing already exists here. Use the existing one, or choose another name or value. |
409 | region_unavailable | The region has no room for this at the moment. Choose another region, or try again later. |
422 | invalid | Fix the field named in field; errors says why. |
422 | runtime_not_adopted | A Node.js version cannot be chosen for a new site here yet. Leave it out to use the default. |
429 | rate_limited | Wait for the seconds in Retry-After. |
502 | none | Read the message; it says what happened. |
503 | temporarily_unavailable | The Control Panel could not answer just now. Nothing was started; try again shortly. |
Reasons are listed by name in Errors and refusals.
The same task elsewhere
curl -X POST https://manage.corefinity.com/api/corestack/clusters \
-H "Authorization: Bearer $COREFINITY_TOKEN" \
-H "Accept: application/json" \
-H "Content-Type: application/json" \
-d '{
"compute_family": "general",
"name": "Northwind UK 2",
"package": "xsmall",
"region": "uk"
}'use Illuminate\Support\Facades\Http;
$response = Http::withToken(env('COREFINITY_TOKEN'))
->acceptJson()
->post('https://manage.corefinity.com/api/corestack/clusters', [
'compute_family' => 'general',
'name' => 'Northwind UK 2',
'package' => 'xsmall',
'region' => 'uk',
]);
$data = $response->json();const res = await fetch('https://manage.corefinity.com/api/corestack/clusters', {
method: 'POST',
headers: {
Authorization: `Bearer ${process.env.COREFINITY_TOKEN}`,
Accept: 'application/json',
'Content-Type': 'application/json',
},
body: JSON.stringify({
compute_family: 'general',
name: 'Northwind UK 2',
package: 'xsmall',
region: 'uk',
}),
});
// An answer with no body (204) has nothing to parse.
const text = await res.text();
const data = text && JSON.parse(text);import os
import requests
res = requests.post(
"https://manage.corefinity.com/api/corestack/clusters",
headers={
"Authorization": f"Bearer {os.environ['COREFINITY_TOKEN']}",
"Accept": "application/json",
},
json={
"compute_family": "general",
"name": "Northwind UK 2",
"package": "xsmall",
"region": "uk",
},
)
data = res.json() if res.content else None{
"cluster": {
"actions": {
"create_environment": {
"available": false,
"reason": "This cluster is still being created"
},
"create_site": {
"available": false,
"reason": "This cluster is still being created"
},
"delete": {
"available": false,
"reason": "This cluster is still being created"
},
"upgrade_package": {
"available": false,
"downgrade": false,
"reason": "This cluster is still being created"
}
},
"allowance": {
"cpu_millicores": 4000,
"disk_bytes": 214748364800,
"memory_bytes": 17179869184
},
"capacity": {
"basis": "estimate",
"cpu": {
"capacity_millicores": 4000,
"level": "ok",
"percent": 0,
"used_p95_millicores": 0
},
"evaluated_at": null,
"level": "ok",
"memory": {
"capacity_bytes": 17179869184,
"level": "ok",
"percent": 0,
"used_p95_bytes": 0
},
"package": {
"cores": 4,
"key": "xsmall",
"memory_bytes": 17179869184,
"name": "Extra small"
},
"sites": {
"count": 0,
"environments": 0
},
"storage": {
"capacity_bytes": 214748364800,
"level": "ok",
"percent": 0,
"provisioned_bytes": 0
}
},
"created_at": "2026-11-14T10:00:00Z",
"created_by": {
"id": 2041,
"name": "Eleanor Finch"
},
"environments": 0,
"id": 1018,
"kind": "customer",
"label": "Northwind UK 2",
"location": {
"arch": "arm64",
"label": "London (europe-west2)"
},
"logs": null,
"name": "Northwind UK 2",
"operation": {
"id": 5130,
"scope": "cluster",
"status": "committed",
"status_label": "Waiting for the platform",
"summary": "New cluster",
"to_version": null,
"type": "cluster_create",
"type_label": "New cluster"
},
"package": {
"cores": 4,
"key": "xsmall",
"name": "Extra small"
},
"pending_changes": {
"count": 0,
"window_starts_at": null
},
"ready_at": null,
"region": {
"key": "uk",
"label": "United Kingdom",
"place": "London (europe-west2)"
},
"reserved": {
"cpu_limit_millicores": 0,
"cpu_request_millicores": 0,
"disk_bytes": 0,
"memory_limit_bytes": 0,
"memory_request_bytes": 0
},
"sites": {
"applying": 0,
"degraded": 0,
"down": 0,
"environments": 0,
"healthy": 0,
"live": 0,
"preview": 0,
"total": 0,
"unknown": 0
},
"slug": "northwind-uk-2",
"snapshots": {
"frequency": "6h",
"over_quota": false,
"pruned_recently": false,
"quota_bytes": 429496729600,
"retention_days": 30,
"usage_bytes": null
},
"ssh_host": "ssh-northwind-uk-2.example.net",
"state": "creating",
"state_label": "Creating",
"state_reason": null,
"synced": {
"inventory_at": "2026-11-14T10:00:00Z",
"live_at": null,
"usage_at": null
},
"usage": {
"cpu_millicores": null,
"disk_used_bytes": null,
"memory_bytes": null,
"observed_at": null,
"source": null
}
},
"operation": {
"applied_at": null,
"can_retry": false,
"cluster": {
"id": 1018,
"name": "Northwind UK 2"
},
"committed_at": "2026-11-14T10:00:01Z",
"completed_at": null,
"created_at": "2026-11-14T10:00:00Z",
"environment": null,
"error": null,
"failure_reason": null,
"from_version": null,
"hot": false,
"id": 5130,
"install": null,
"label": "SSH gateway ready",
"notes": [],
"parent_operation_id": null,
"requested_by": {
"id": 2041,
"name": "Eleanor Finch"
},
"result": null,
"result_state": "none",
"retried_by_operation_id": null,
"retries_operation_id": null,
"retry_requires": null,
"scope": "cluster",
"site": null,
"status": "committed",
"status_label": "Publishing the SSH gateway",
"step": "gateway_ready",
"steps": [
{
"key": "requested",
"label": "Requested",
"state": "done"
},
{
"key": "gateway_ready",
"label": "SSH gateway ready",
"state": "current"
},
{
"key": "ready",
"label": "Ready",
"state": "todo"
}
],
"subject": null,
"summary": "New cluster",
"to_version": null,
"type": "cluster_create",
"type_label": "New cluster"
},
"sites": []
}{
"cluster": {
"actions": {
"create_environment": {
"available": false,
"reason": "This cluster is still being created"
},
"create_site": {
"available": false,
"reason": "This cluster is still being created"
},
"delete": {
"available": false,
"reason": "This cluster is still being created"
},
"upgrade_package": {
"available": false,
"downgrade": false,
"reason": "This cluster is still being created"
}
},
"allowance": {
"cpu_millicores": 4000,
"disk_bytes": 214748364800,
"memory_bytes": 17179869184
},
"capacity": {
"basis": "estimate",
"cpu": {
"capacity_millicores": 4000,
"level": "ok",
"percent": 0,
"used_p95_millicores": 0
},
"evaluated_at": null,
"level": "ok",
"memory": {
"capacity_bytes": 17179869184,
"level": "ok",
"percent": 0,
"used_p95_bytes": 0
},
"package": {
"cores": 4,
"key": "xsmall",
"memory_bytes": 17179869184,
"name": "Extra small"
},
"sites": {
"count": 0,
"environments": 0
},
"storage": {
"capacity_bytes": 214748364800,
"level": "ok",
"percent": 0,
"provisioned_bytes": 0
}
},
"created_at": "2026-11-14T10:00:00Z",
"created_by": {
"id": 2041,
"name": "Eleanor Finch"
},
"environments": 0,
"id": 1018,
"kind": "customer",
"label": "Northwind UK 2",
"location": {
"arch": "arm64",
"label": "London (europe-west2)"
},
"logs": null,
"name": "Northwind UK 2",
"operation": {
"id": 5130,
"scope": "cluster",
"status": "pending",
"status_label": "Saving change",
"summary": "New cluster",
"to_version": null,
"type": "cluster_create",
"type_label": "New cluster"
},
"package": {
"cores": 4,
"key": "xsmall",
"name": "Extra small"
},
"pending_changes": {
"count": 0,
"window_starts_at": null
},
"ready_at": null,
"region": {
"key": "uk",
"label": "United Kingdom",
"place": "London (europe-west2)"
},
"reserved": {
"cpu_limit_millicores": 0,
"cpu_request_millicores": 0,
"disk_bytes": 0,
"memory_limit_bytes": 0,
"memory_request_bytes": 0
},
"sites": {
"applying": 0,
"degraded": 0,
"down": 0,
"environments": 0,
"healthy": 0,
"live": 0,
"preview": 0,
"total": 0,
"unknown": 0
},
"slug": "northwind-uk-2",
"snapshots": {
"frequency": "6h",
"over_quota": false,
"pruned_recently": false,
"quota_bytes": 429496729600,
"retention_days": 30,
"usage_bytes": null
},
"ssh_host": "ssh-northwind-uk-2.example.net",
"state": "creating",
"state_label": "Creating",
"state_reason": null,
"synced": {
"inventory_at": "2026-11-14T10:00:00Z",
"live_at": null,
"usage_at": null
},
"usage": {
"cpu_millicores": null,
"disk_used_bytes": null,
"memory_bytes": null,
"observed_at": null,
"source": null
}
},
"operation": {
"applied_at": null,
"can_retry": false,
"cluster": {
"id": 1018,
"name": "Northwind UK 2"
},
"committed_at": null,
"completed_at": null,
"created_at": "2026-11-14T10:00:00Z",
"environment": null,
"error": null,
"failure_reason": null,
"from_version": null,
"hot": false,
"id": 5130,
"install": null,
"label": "Requested",
"notes": [],
"parent_operation_id": null,
"requested_by": {
"id": 2041,
"name": "Eleanor Finch"
},
"result": null,
"result_state": "none",
"retried_by_operation_id": null,
"retries_operation_id": null,
"retry_requires": null,
"scope": "cluster",
"site": null,
"status": "pending",
"status_label": "Preparing the cluster",
"step": "requested",
"steps": [
{
"key": "requested",
"label": "Requested",
"state": "current"
},
{
"key": "gateway_ready",
"label": "SSH gateway ready",
"state": "todo"
},
{
"key": "ready",
"label": "Ready",
"state": "todo"
}
],
"subject": null,
"summary": "New cluster",
"to_version": null,
"type": "cluster_create",
"type_label": "New cluster"
},
"sites": []
}{
"message": "Unauthenticated."
}{
"code": "company_unavailable",
"message": "Your company is unavailable.",
"reason": "company_unavailable"
}{
"code": "feature_off",
"message": "Not found.",
"reason": "feature_off"
}{
"code": "cluster_busy",
"message": "The package or the region changed while the request was checked. Check it and try again.",
"reason": "cluster_busy"
}{
"code": "invalid",
"errors": {
"sites.0.tier": [
"Choose a size"
]
},
"field": "sites.0.tier",
"message": "Choose a size",
"reason": "invalid"
}{
"code": "rate_limited",
"message": "Too many requests. Try again in 60 seconds.",
"reason": "rate_limited",
"retry_after_seconds": 60
}{
"message": "The platform could not be checked just now. Try again in a minute."
}{
"code": "temporarily_unavailable",
"message": "This region is not ready",
"reason": "temporarily_unavailable"
}Try it is off for this endpointIt can add to your bill. Copy the request and run it yourself when you are sure.