Hi Yukun,
I think you've found a bug in the code. The accumulators don't seem to be
really thread safe. I've created an issue to fix this issue [1]. Thanks for
reporting the problem :-)
[1] https://issues.apache.org/jira/browse/FLINK-4829
Cheers,
Till
On Fri, Oct 14, 2016 at 8:32 AM, Yukun Guo
This happens when the TaskManager is serializing an
org.apache.flink.api.common.accumulators.Histogram by iterating through the
underlying TreeMap while a MapFunction for updating the accumulator
attempts to modify the TreeMap concurrently. How could I fix it?
The call stack:
WARN org.apache.fl
