Hi Steve,
RocksDB does not contribute to the JVM direct memory. RocksDB off-heap
memory consumption is part of managed memory [1].
You got `OutOfMemoryError: Direct buffer memory` which is related to
the JVM direct memory, also off-heap but managed by JVM.
The JVM direct memory limit depends on t
We recently migrated to Flink 1.10, but are experiencing some issues with
memory.
Our cluster is:
1) Running inside of Kubernetes
2) Running in HA mode
3) Checkpointing/Savepointing to an HDFS cluster inside of Kubernetes
4) Using RocksDB for checkpointing
5) Running on m5d.4xlarge EC2 instances w