WebSocket notification stream dropping Conversation events after 200ms burst

Trying to subscribe to routing:conversation:activity events via the WebSocket endpoint /api/v2/notifications/streams. The connection establishes fine, but I’m only receiving about 10% of the expected events when there’s a high volume of interactions.

Here’s the subscription payload I’m sending:

{
 "eventSubscriptions": [
 {
 "eventType": "routing:conversation:activity",
 "filters": {
 "queueId": "abc-123"
 }
 }
 ]
}

I’m using the Python SDK genesyscloud.platformclient to handle the connection. The server sends a keep-alive ping, but it seems like the batch size or the reconnection logic is swallowing events. I’ve increased the buffer size in the client, but that didn’t help.

Environment specs:

  • Python 3.11
  • Genesys Cloud SDK v3.14
  • Queue with ~500 concurrent chats
  • No custom filters other than queueId

I’ve checked the Jaeger traces for the client side, and the spans show the WebSocket connection is active, but the event handler isn’t being triggered for most events. Is there a rate limit on the server side for WebSocket streams? Or is there a specific header I need to set to increase the throughput? I’ve tried adding X-Genesys-Client-Id but no change.

Any ideas on how to debug this? I can’t seem to find the dropped events in the logs.