logstash:Connection reset by peer==>filebeat:Failed to publish events

问题描述：filebeat-->logstash-->elasticsearch-->kibana这种架构。
问题：logstash-->到es
[ERROR][org.logstash.beats.BeatsHandler] Exception: Connection reset by peer 
有这累的报错后，
filebeat--> 到logstash
ERR Failed to publish events caused by: read tcp xxxx:47752->xxxx:5516: i/o timeout
ERR Failed to publish events caused by: client is not connected
然后就开始停止发送数据了
INFO No non-zero metrics in the last 30s
INFO No non-zero metrics in the last 30s
INFO No non-zero metrics in the last 30s
INFO No non-zero metrics in the last 30s
INFO No non-zero metrics in the last 30s
 
 
请问大神有没有碰到过类似的问题的？是怎么解决的？
现在基本状况是：
es filebeat都做过优化，目前es是一台机器2实例，4分片0副本，索引按天算，一天500多G，有5个logstash进程往es写入数据，5台机器的10个filebeat往logstash发送数据。