tzulitai opened a new pull request #142: URL: https://github.com/apache/flink-statefun/pull/142
This PR adds in total 2 new metrics related to backpressure: - Number of blocked addresses (per function type) - Number of inflight async operations (per function type + per operator) We also rejected to add the following metric, since after some discussion it doesn't seem to add much value: - Number of accumulated records pre-flight in batch per function type. This was not added, with the assumption that users would really only want to care about that some address has reached the maximum request batch size and was being blocked. --- ## Verification I verified this by running the Python Greeter example, with the following modifications to let backpressure happen more easily: - Maximum batch size = 1 - No delay between each generated message to have maximum input rate You can see the following metric charts in the Flink Web UI:  --- ## Brief changelog - 78cbc19 Extends the `FunctionTypeMetrics` interface to include the new metrics, and adds a new `FunctionDispatcherMetrics` interface for per-operator metrics. - 6408866 Introduce a scoped-down interface `FunctionTypeMetricsRepository` and let `StatefulFunctionsRepository` extend it. Components that need to access function metrics will be passed this interface. - 0fbb17a preliminary extension to `ObjectContainer` DI utility so that we can share same instance across different object labels. - 6c77ca4 Wire-in the new metrics in `AsyncSink` to expose backpressure metrics. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
