missed heartbeats from client, timeout: 60s

[段杨龙]

In the rabbitmq three-node cluster in openstack, one of the nodes is shut down, and the error-level closing AMQP connection frequently occurs after the other two nodes detect that the other node is down, because missed heartbeats from client, timeout: 60s, and these error logs occur There is no corresponding accepting AMQP connection log for the corresponding port. I want to know the reason.