If you see this behavior, it indicates that ICMP echo requests sent from the BE TEA Agent machine to the BE worker machine are being blocked. To confirm, login to the BE TEA Agent machine, and attempt to ping the IP address of the BE worker machine: $ ping ec2-54-89-221-208.compute-1.amazonaws.com PING ec2-54-89-221-208.compute-1.amazonaws.com (54.89.221.208) 56(84) bytes of data. ^C --- ec2-54-89-221-208.compute-1.amazonaws.com ping statistics --- 5 packets transmitted, 0 received, 100% packet loss, time 4162ms
The above example shows 100% packet loss, and no successful pings. You can also enable debug-level logging for the BE TEA Agent, and look for messages like.. 2022 Feb 21 12:01:11.016 GMTZ DEBUG host-pinger [service.impl.BEMasterHostManagementServiceImpl] - machine0 is Unreachable
To resolve, ensure that inbound ICMP echo requests, sent from the BE TEA Agent machine, can be received on the BE worker machine. Consult with your IT team as needed to confirm any changes you may need to make with respect to firewalls or other security settings that are relevant for your machines. Once ICMP echo requests are enabled, the status should change to "reachable":
NOTE: If your IT team will not allow echo requests, deployments will still succeed if you have properly configured SSH (tcp port 22) and JMX (tcp port 5500, by default) between the BE TEA Agent machine and the BE worker machine.
For details on how to configure SSH key-based authentication (i.e. "password-less authentication"), refer to the Knowledge article: "How to configure SSH key-based authentication between the BE TEA Agent and a BE Worker machine".
For details on how to configure JMX communication, refer to the Knowledge article: "BusinessEvents application instance can be started from TEA, but cannot be monitored in TEA".