I'm using ActiveMQConnectionFactory from ActiveMQ Classic 5.16.4 with the TCP protocol from my JMS client. The JMS broker is ActiveMQ Classic 5.16.5. No matter what parameters I can think of I cannot seem to prevent or delay the client from disconnecting. Using the failover layer does functionally help, but the environment settings result in a log message every time they reconnect, and with 8+ clients at once, it's a ton of bloat.
I've tried looking at the tcp transport reference (as well as the source code to see if there were any other settings I could use), and none of the settings have helped. I've tried keepAlive set to true, soTimeout set to 0, soWriteTimeout set to 0, and connectionTimeout set to 0.
I'm formatting my URIs in the form of tcp://hostname:XXXXX?connectionTimeout=0&keepAlive=true, which is correct as far as I'm aware.
The network situation is inherently obscenely complicated due to the nature of this work. The old scenario is both client and broker are on the same VM, and did not have issues with disconnecting, nor had logging overflowing when failover was enabled. The current scenario, the one with the issue described above, is with the broker in a Kubernetes container.
EDIT: I should probably mention the broker URL: tcp://hostname:XXXXX?maximumConnections=1000&wireFormat.maxFrameSize=104857600&transport.useInactivityMonitor=false.