Webhook Strategy (ASYNC)

Rate Limit Policy

All OEM Alerting API endpoints are subject to punitive rate limiting. Limits are applied
per source IP address within a sliding time window.

Rate limits may vary by environment and contract. Contact your MSPCentric
account representative for your specific allocation.

Rate Limit Response

When a rate limit is exceeded, the API returns:

• HTTP Status: 429 Too Many Requests
• Header: Retry-After: <seconds> -- the number of seconds until the rate
  limit window resets
• Body:
  JSON

  {
    "statusCode": 429,
    "message": "Rate limit exceeded. Retry after <seconds> seconds."
  }

Recommended Client Behavior

1. Respect Retry-After -- Pause requests for the duration specified in the
   Retry-After header before retrying.
2. Implement exponential backoff -- If consecutive 429 responses are received,
   increase the wait time exponentially (e.g., 1s, 2s, 4s, 8s) up to a reasonable
   maximum.
3. Queue locally -- For high-volume alert sources, maintain a local queue and
   submit alerts at a controlled rate within your allocation.

Ingestion

Ingestion Latency

Upon successful ingestion, the API returns:

{
  "statusCode": 201,
  "message": "Alert accepted for processing",
  "data": {
    "correlationId": "abc-123-def-456",
    "messageId": "sqs-message-id"
  }
}

The correlationId can be used to query the alert status at any point via the
/alert/:correlationId/ticket endpoint.

Durability

Accepted alerts are persisted to a durable message queue before the
201 response is returned. Messages are retained for up to 14 days in the event of
processing failures.

Failed messages are routed to a dead-letter queue after the configured retry
policy is exhausted.

Processing

Processing Time

Processing time is measured from queue ingestion to PSA ticket action completion.

Important: Processing time is dependent on the responsiveness of the downstream
PSA system. Degradation of the PSA API may impact
processing times beyond MSPCentric's control.

Processing Throughput

Alert processing capacity scales with infrastructure. Current capacity targets:

Throughput targets assume typical PSA API latency (~500ms). Actual throughput may
vary based on PSA performance and alert complexity.

5.4 Status Delivery

Status updates are delivered via the Inform system: