Hi,
We are experiencing high cpu usage when the cypher-shell process initiated by users are not closed properly because of unix session timeout, please can you help fixing this.
Is there a way we can set idle session timeout at client end?
Linux Machines: AWS ec2 instances with RHEL 8.
Neo4j: 4.4.3
Issue Scenario:
User establish a ssh connection from jump host to the AWS ec2 instance and use cypher command to connect to the neo4j DB.
Once AWS ec2 ssh/ssm connectivity is closed the cypher-shell process remains active and started to consume more cpu resource (one complete core), more than one such process started to saturate cpu resource(cores).
====
[root@server-1 ~]# ps -def|grep cypher
root 590136 590108 0 11:02 pts/2 00:00:02 /usr/bin/java -jar /usr/bin/../share/cypher-shell/lib/cypher-shell.jar
root 590280 590227 1 11:03 pts/4 00:00:01 /usr/bin/java -jar /usr/bin/../share/cypher-shell/lib/cypher-shell.jar
===
[root@server-1 ~]# netstat -anplt|head -2;netstat -anplt|grep "590136"
Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name
tcp6 0 0 127.0.0.1:49306 127.0.0.1:7687 ESTABLISHED 590136/java
tcp6 0 0 127.0.0.1:49304 127.0.0.1:7687 ESTABLISHED 590136/java
[root@server-1 ~]# netstat -anplt|head -2;netstat -anplt|grep "590280"
Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name
tcp6 0 0 127.0.0.1:49380 127.0.0.1:7687 ESTABLISHED 590280/java
tcp6 0 0 127.0.0.1:49378 127.0.0.1:7687 ESTABLISHED 590280/java
===
Thank you,
ruben.