I have the same problem, even more impactful. Some subtasks stall forever quite
consistently.
I am using Flink 1.7.1, but I've tried downgrading to 1.6.3 and it didn't help.
The Backend doesn't seem to make any difference, I've tried Memory, FS and
RocksDB back ends but nothing changes. I've also
or RETAIN_ON_CANCELLATION. I assume that using the
MemoryStateBackend and CANCELLATION should remove any possibility of disk/IO
congestions, am I wrong?.
Pasquale
From: Till Rohrmann
Sent: 15 January 2019 10:33
To: Pasquale Vazzana
Cc: Bruno Aranda ; user
Subject: Re: Subtask much slower than
