We have a cluster of three nodes. Two of them are on one server, the third node is on the other. All queues have an "ha-all" availability setting.
We ran into a problem that sometimes the cluster stops receiving messages, or receives them very slowly. Some of the connections are sometimes marked "flow". All nodes are not lightly loaded (~10% CPU, ~7% RAM). In addition, it was noticed that at such moments the "rabbitmqctl cluster_status" request hangs on all nodes and responds only after 10-15 minutes. At the same time, the connection between servers and nodes is not lost (ping is stable ~6ms)
Also, during such problems, the ManagementUI freezes or takes a very long time to respond to requests.
What could be the problem?
