Delay in Web interface responses when a node is down or stopped

[pavel-semenchukov]

Good evening, I have a problem. I am configuring a cluster of 3 nodes. Everything works as I want, except for one moment, if at least one node stops working, I have very big delays when using the web interface, switching to Queues and Streams or Exchanges or Overview tabs happens with a very big delay, 20-30 seconds. Can you tell me who has encountered this, how did you manage to solve it? I have tried many different parameters in the config, but I was not able to change this behavior

[lukebakken]

We can't assist you without more information:

• RabbitMQ version
• Erlang version
• Operating system running RabbitMQ, and version

[pavel-semenchukov]

I have
RabbitMQ 3.12.12
Erlang 26.2.1
Operating System: Ubuntu 22.04.3 LTS

[pavel-semenchukov]

I'm currently setting up the cluster for training purposes, and there is no real client load on it.

[michaelklishin]

This has been discussed several times fairly recently:

• Slow operations in degraded cluster #10100
• Management interface (rabbitmq_management plugin) timing out when a node is offline #9522
• Optimise HTTP API /queues endpoint #9874 (will ship in 3.13.0)
• Presence server #10065 (will ship in 4.0)

There isn't much else to add, really.