Performance issues when RocksDB block cache is full

2021-02-13 Thread Yaroslav Tkachenko
Hello, I observe throughput degradation when my pipeline reaches the maximum of the allocated block cache. The pipeline is consuming from a few Kafka topics at a high rate (100k+ rec/s). Almost every processed message results in a (keyed) state read with an optional write. I've enabled native Roc

Re: How to debug flink serialization error?

2021-02-13 Thread Debraj Manna
Thanks Robert for the pointers. It is some issue with mockito which I was using to mock getMetricStoreProgramHelper method in my unit test. For now I have modified my unit test to not use mockito. I will try to provide a reproducible example. On Fri, Feb 12, 2021 at 8:56 PM Robert Metzger wrote