Developing a real-time dashboard for a premium AppFoundry integration, the application requires frequent polling of the Analytics API to display live interaction metrics. The endpoint in question is GET /api/v2/analytics/interactions/summary.
Despite implementing standard exponential backoff and respecting the documented rate limits, the application consistently receives 429 Too Many Requests responses during peak hours. The request headers include the correct X-Genesys-Client-Id and valid OAuth tokens for a multi-tenant environment. The frequency is set to 15-second intervals per tenant, which should theoretically fall within the limits for a Premium app tier.
Has anyone encountered specific throttling behaviors for Analytics endpoints that differ from the standard Platform API limits? The error payload does not provide a Retry-After header, making it difficult to implement dynamic backoff. Checking if there are undocumented per-tenant limits or if the X-Genesys-Client-Id header is being ignored in the rate-limiting logic for analytics-specific endpoints. Any insights into optimizing this or alternative approaches for high-frequency data retrieval would be appreciated.