问题
I see the below error in worker logs, it happens almost every milliseconds, but the cluster is running fine, I wanted to know what does these error mean and any idea on why this would occur.
This happens on all the worker nodes
2016-05-12T15:32:53.514-0500 b.s.m.n.Client [ERROR] connection attempt 3 to Netty-Client-xxxxx.hq.abc.com/xx.xx.xxx.xx:6700 failed: java.net.ConnectException: Connection refused: xxxxx.hq.abc.com/xx.xx.xxx.xxx:6700
And after some time i see this 2016-05-12T15:44:25.940-0500 b.s.m.n.Client [ERROR] discarding 1 messages because the Netty client to Netty-Client-xxxxx.hq.abc.com/xx.xx.xxx.xxx:6700 is being closed
回答1:
After struggling forever with this problem, I found that setting storm.local.hostname property in the storm.yaml file solved the issue for me. On my laptop, I set storm.zookeeper.servers, nimbus.host and storm.local.hostname all to "localhost".
I am using version 0.10.2 of storm.
来源：https://stackoverflow.com/questions/37197419/connection-refused-error-in-worker-logs-apache-storm