Persistence user on rabbitmq

[ghost]

So I was make my rabbitmq and I noticed that, when I persiste the volume, my users don't save when I make docker-compose down and up, and I saw that, my rabbitmq creates a new "cluester" all time that I make the docker compose up, so how can I stay with a one cluster, have any thing that can I to do to pin just one cluster name, to use one cluster with the save config

[michaelklishin]

@YukioBunny you already asked exactly the same question about node's data directory persistence in #605 and was offered recommendations. Nodes only initialize new clusters if they do not have an existing data directory to use (because the volume is not persistent or the data directory path has changed).

This will involve log messages such as

2023-02-06 15:56:40.145883-05:00 [info] <0.229.0> Node database directory at /tmp/123 is empty. Assuming we need to join an existing cluster or initialise from scratch...

2023-02-06 15:56:40.221024-05:00 [info] <0.229.0> Initialising internal cluster ID to 'rabbitmq-cluster-id-IDN4yCLL1zZINr2hifW8pw'

in node logs.

And then they perform peer discovery on the first boot, and log the result.

[yosifkit]

You likely need to provide a consistent hostname: https://github.com/docker-library/docs/tree/bb6b29b9a31fe8fe9e083c989b3091914e301fd6/rabbitmq#running-the-daemon