We harvest logs on aws machines by filebeat. Filebeat on each machine push those log to local logstash nodes and then logstash nodes push the data to local Elasticsearch nodes. It seems to me that both logstash node and Elasticsearch nodes have no of tcp connection limit. Each logstash node cannot accept more than around 100 tcp connections with filebeat in aws machines. Something like that also happen on Elasticsearch nodes, each ES node cannot form more than around 100 tcp connections with logstash node. Timeout error will occurs in tcp connection exceed around 100 or more. Is there any way to increase no of tcp connection? is it a parameter of jvm or linux itself?
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.