Endpoint reference

KernelHost API

All endpoints return application/json. Errors follow RFC 7807 (Problem+JSON). Every response carries an X-Request-Id header for correlation with the server-side audit log.

MethodPathScopePurpose
GET/v1/healthpublicUnauthenticated health check.
GET/v1/meanyOwn account info + key metadata.
GET/v1/productsread:productsList of all publicly listed products with prices.
GET/v1/products/{id}read:productsProduct details including config options.
POST/v1/orderswrite:ordersPlace an order. Idempotency-Key required.
GET/v1/orders/{id}read:ordersStatus of an own order.
GET/v1/servicesread:servicesList of own services (paginated).
GET/v1/services/{id}read:servicesDetails of an own service.
GET/v1/services/{id}/credentialsread:credentialsService credentials (separate scope, audit-logged).
POST/v1/services/{id}/actionswrite:servicesService action: start, stop, reboot, cancel (at the end of the billing period), cancel_revoke.
GET/v1/services/{id}/metricsread:servicesServer metrics of an own service (CPU, memory, network).
GET/v1/services/{id}/ddos/profilesread:ddosAvailable Advanced DDoS Protection profiles.
GET/v1/services/{id}/ddos/rulesread:ddosCurrent rules for the assigned protection IP.
POST/v1/services/{id}/ddos/ruleswrite:ddosAdd one protection rule. Idempotency-Key required.
DELETE/v1/services/{id}/ddos/rules/{rule_id}write:ddosDelete one owned rule. Idempotency-Key required.
GET/v1/services/{id}/ddos/attacksread:ddosAttacks detected on the protected IP (Remote DDoS Protection).
GET/v1/services/{id}/ddos/attacks/{attack_id}read:ddosDetails of one attack: traffic and breakdowns.
GET/v1/billing/balanceread:billingCredit balance.
GET/v1/billing/invoicesread:billingList of your invoices (paginated).
GET/v1/billing/invoices/{id}/pdfread:billingInvoice PDF.
GET/v1/webhooksread:webhooksCurrent webhook URL.
PUT/v1/webhookswrite:webhooksSet or unset webhook URL.

Advanced DDoS Protection rules

These endpoints require an Advanced DDoS Protection service owned by your account. Reads work in active or suspended status; changes require active status. Enable read:ddos for reads and write:ddos for changes explicitly on your API key. Existing keys receive no new permissions automatically. Tunnel and whitelist management are unavailable.

Choose app_code from the profile catalog. port must be an integer from 1 to 65535; for icmp omit it or send 0. Optional extra_ports maps exact profile labels to integer ports; omitted labels use their profile defaults. Protocols come from the server, so do not send protocol. Unknown fields or labels return HTTP 400.

GET /v1/services/{id}/ddos/profiles

{
  "data": [{
    "app_code": "rust",
    "name": "Rust",
    "default_port": 28015,
    "protocol": "UDP",
    "extra_ports": [
      {"name": "Query", "port": 28016, "protocol": "UDP"},
      {"name": "RCON", "port": 28016, "protocol": "TCP"},
      {"name": "RustPlus", "port": 28082, "protocol": "TCP"}
    ]
  }],
  "meta": {"service_id": 1234, "protection_ip": "192.0.2.10", "max_rules": 1000}
}

POST /v1/services/{id}/ddos/rules

{
  "app_code": "rust",
  "port": 28015,
  "extra_ports": {"Query": 28016, "RCON": 28016, "RustPlus": 28082}
}

POST returns HTTP 201 with the created rule in data. GET returns the rules and meta with service_id, protection_ip and max_rules. DELETE takes an empty body and returns HTTP 200 as shown below. POST and DELETE both require Idempotency-Key.

GET /v1/services/{id}/ddos/rules

{
  "data": [{
    "id": "example-rule-id",
    "app_code": "rust",
    "port": 28015,
    "protocol": "UDP",
    "extra_ports": [
      {"name": "Query", "port": 28016, "protocol": "UDP"},
      {"name": "RCON", "port": 28016, "protocol": "TCP"},
      {"name": "RustPlus", "port": 28082, "protocol": "TCP"}
    ]
  }],
  "meta": {"service_id": 1234, "protection_ip": "192.0.2.10", "max_rules": 1000}
}

DELETE /v1/services/{id}/ddos/rules/{rule_id}

{"data": {"id": "example-rule-id", "deleted": true}}

Panel and API share fixed UTC windows: 10 write attempts per minute per account, 5 per minute per protection IP and 200 per UTC day per account. Admitted invalid or failed attempts count; successful idempotency replays do not. DDoS API reads share 30 requests per minute per account. Maximum: 1000 rules per protection IP.

HTTP 404: unknown or foreign service, or foreign, missing or system rule. HTTP 409: not an Advanced DDoS Protection service (not_supported_for_product), inactive service, port collision or concurrent change. HTTP 429: limit reached, respect Retry-After. HTTP 502: protection service could not complete the request. The system block rule is hidden and cannot be deleted. HTTP 503: request limits are temporarily unavailable; respect Retry-After.

DDoS attack reports

These endpoints require a Remote DDoS Protection service of your account in active or suspended status and the scope read:ddos, which must be enabled explicitly. The list shows the DDoS attacks detected on the protected IP address of the service, newest first.

Query parameters: from and to as a date (YYYY-MM-DD) or an ISO 8601 date-time with time zone (default: the last 30 days), limit from 1 to 100 (default 25) and offset from 0 to 9999 (default 0). The window never reaches back more than 90 days and never before the protected IP address was assigned to this service; meta.from and meta.to show the window actually used.

GET /v1/services/{id}/ddos/attacks?from=2026-10-10

{
  "data": [{
    "id": "3f6c1a2e-9b47-4d8a-b5e2-7c0d9e4a1f63",
    "type": "ddos_attack",
    "started_at": "2026-11-07T18:34:06Z",
    "ended_at": null,
    "detected_bps": 20938227712,
    "detected_pps": 1998848,
    "source_ip_count": 341,
    "target_ip": "203.0.113.111",
    "details_available": true
  }],
  "paging": {"limit": 25, "offset": 0, "total": 1, "truncated": false},
  "meta": {"service_id": 5678, "protected_ip": "203.0.113.111", "from": "2026-10-10T00:00:00Z", "to": "2026-11-09T11:10:16Z", "retention_days": 90}
}

Each attack has id (UUID), type (ddos_attack, or blackhole if the target IP was temporarily null-routed), started_at, ended_at (can be null), detected_bps, detected_pps, source_ip_count, target_ip and details_available. detected_bps and detected_pps are the rate at the moment of detection; filtering had already started then, so they can be lower than the peak in the details. paging contains limit, offset, total and truncated (true if the window holds more than 500 attacks), meta contains service_id, protected_ip, from, to and retention_days (90). Lists are cached for 60 seconds.

GET /v1/services/{id}/ddos/attacks/{attack_id}

{
  "data": {
    "id": "3f6c1a2e-9b47-4d8a-b5e2-7c0d9e4a1f63",
    "type": "ddos_attack",
    "started_at": "2026-11-07T18:34:06Z",
    "ended_at": null,
    "detected_bps": 20938227712,
    "detected_pps": 1998848,
    "source_ip_count": 341,
    "target_ip": "203.0.113.111",
    "details_available": true,
    "peak_bps": 20938227712,
    "peak_pps": 1998848,
    "total_bytes": 104809201664,
    "total_packets": 81887232,
    "interval_seconds": 1,
    "traffic": [{"time": "2026-11-07T18:34:06Z", "bps": 20938227712, "pps": 1998848}],
    "protocols": [{"protocol": "UDP", "avg_bps": 695618765, "avg_pps": 65645, "bytes": 104342814720, "packets": 78774272, "bps_percent": 99.56, "pps_percent": 96.2}],
    "destination_ports": [{"port": 58675, "protocol": "UDP", "avg_bps": 695556506, "avg_pps": 65481, "bps_percent": 99.55, "pps_percent": 95.96}],
    "source_ports": [{"port": 53, "protocol": "UDP", "avg_bps": 41896312, "avg_pps": 3511, "bps_percent": 5.996, "pps_percent": 5.15}],
    "source_countries": [{"country": "MX", "avg_bps": 96051528, "avg_pps": 8656}],
    "source_ips": [{"ip": "198.51.100.7", "avg_bps": 46162248, "avg_pps": 4151, "bps_percent": 6.61, "pps_percent": 6.08}],
    "packet_sizes": [{"range": "1400-1499", "avg_bps": 374806938, "percent": 54.076}]
  },
  "meta": {"service_id": 5678, "cached": false}
}

The details contain all list fields plus peak_bps, peak_pps, total_bytes, total_packets and interval_seconds. traffic lists the incoming attack traffic per interval ({time, bps, pps}) as seen by the filtering network, from about 10 minutes before to about 10 minutes after detection. The breakdowns protocols, destination_ports, source_ports, source_countries, source_ips and packet_sizes usually contain the top 5 entries: avg_bps and avg_pps are averages over the measurement window, bps_percent and pps_percent are shares. protocols also contains bytes and packets, packet_sizes contains range, avg_bps and percent. Details are cached for up to 15 minutes.

HTTP 400 validation_failed: invalid value or unknown query parameter. HTTP 404: unknown or foreign service, unknown attack ID or attack on another IP address. HTTP 409 not_supported_for_product: not a Remote DDoS Protection service (for example Advanced DDoS Protection or a server). HTTP 409 service_not_active: the service is neither active nor suspended. HTTP 409 protection_not_ready: the protection is still being set up.

Server metrics

This endpoint requires an active service of your account and the scope read:services. Edge Unlimited Traffic VPS provide values for CPU, memory and network in and out (bit/s and packets/s), plus outbound traffic per day and month to date. Regular Unlimited Traffic VPS in Europe, Germany, France, Canada East, the United Kingdom and Poland provide network in and out in bit/s only.

Query parameter range: 1h, 6h, 24h, 7d or 14d (default 1h). The step between points depends on the range: for Edge Unlimited Traffic VPS about 15 seconds (1h), 20 seconds (6h), 2 minutes (24h), 10 minutes (7d) and 20 minutes (14d); for regular Unlimited Traffic VPS 5 minutes up to 24h, 30 minutes for 7d and 2 hours for 14d. step_seconds shows the step of the response.

GET /v1/services/{id}/metrics?range=24h

{
  "data": {
    "service_id": 1234,
    "range": "24h",
    "step_seconds": 120,
    "available": ["cpu_percent", "memory_percent", "network_in_bps", "network_out_bps", "network_in_pps", "network_out_pps"],
    "summary": {
      "cpu_percent": {"avg": 3.1, "max": 41.7, "last": 2.9},
      "network_in_bps": {"avg": 64012.5, "max": 912345.0, "last": 60026.2}
    },
    "points": [
      {"time": "2026-11-08T11:12:00Z", "cpu_percent": 3.254, "memory_percent": 43.001, "network_in_bps": 68515.425, "network_out_bps": 650.767, "network_in_pps": 119.992, "network_out_pps": 0.775},
      {"time": "2026-11-08T11:14:00Z", "cpu_percent": 3.118, "memory_percent": 43.004, "network_in_bps": 66120.871, "network_out_bps": 702.134, "network_in_pps": 115.317, "network_out_pps": 0.825}
    ],
    "traffic": {
      "month_to_date_out_bytes": 111844510,
      "daily": [{"date": "2026-11-09", "out_bytes": 10485760, "final": false}],
      "as_of": "2026-11-09T11:00:01Z",
      "unit": "bytes",
      "direction": "outbound"
    }
  },
  "meta": {"generated_at": "2026-11-09T11:12:30Z", "cached": false, "measured_at": "server_network_interface"}
}

Values are measured at the server's network interface, so they show traffic that actually reached the server. Memory needs the guest agent inside the server. Values that could not be measured are null; available lists the metrics with measured values, and summary shows avg, max and last for each of them. traffic contains outbound traffic in bytes per day and month to date (final is true once a day is complete) and is null for products without traffic counters. Responses are cached for 30 seconds (1h) up to 10 minutes (14d); meta.cached shows whether the answer came from the cache.

HTTP 400 validation_failed: other range value or unknown query parameter. HTTP 404: unknown or foreign service. HTTP 409 not_supported_for_product: no metrics available for this service (other products and locations). HTTP 409 service_not_active: the service is not active.

For server metrics and attack reports, the general rate limits apply. In addition, fetches that cannot be answered from the cache are limited to 20 per minute per account across these three endpoints; above that you get HTTP 429 with Retry-After. HTTP 502 monitoring_unavailable or reports_unavailable: the data source is temporarily unavailable, retry later. HTTP 503 temporarily_unavailable with Retry-After: a parallel request is preparing the same data.

Not available via the API: attack reports for Advanced DDoS Protection and for servers such as Edge Unlimited Traffic VPS; statistics of clean or dropped traffic per protected IP; CPU and memory for regular Unlimited Traffic VPS; webhook delivery and CSV export.

Idempotency

POST /v1/orders and POST /v1/services/{id}/actions require an Idempotency-Key header (1 to 80 characters from [A-Za-z0-9_.-]). The key is bound to method, path and body. Repeating the same request with the same key within one hour returns the stored response; after that you get 410 with reason replay_stale and should send a new key. The same key with a different body or on another endpoint is rejected with 409 idempotency_conflict.

Rate limits

By default 60 requests per minute (plus a burst of 20) and 5,000 per day per key, 120 per minute per account across all keys and 600 per minute per source IP. When a limit is exceeded, you get HTTP 429 with a Retry-After header. The response headers X-RateLimit-Remaining and X-RateLimit-Reset show the current state of the key limit.