Open pcornelissen opened 3 years ago
To be more precise why this is a problem:
When you use rabbitMQ with the most common defaults and you use spring cloud bus, then you end up (too) often with failing service restarts, because there is a problem to "reconnect" to the implicitily declared inter-service queues. So you do nothing wrong, nothing broke basicly due to your own mistakes and still your services may end up in a restart loop in kubernetes for example. That is very bad and requires manual intervention. I am not sure whether or not the problem is in RabbitMQ, it's defaults, the spring.cloud bus code, its defaults or my (and others usage of cloud bus. But what seems to help is the suggestion of @thpeng in #73
Hey, I am configuring the remote rabbitmq credentials in spring cloud bus application but in logs I am getting error in creating queue to remote rabbitmq
But when I configuring the local rabbitMq credentials in my application it is working as expected, I. Am not understanding the reason behind this Can anyone help me over this
in issue #73 @thpeng found a config setting which helps that restarting services don fail to start due to weird que problems.
Maybe there should be a visible warning or hint that using the different que_master_locator setting the problem disappears.
(The issue is quite serious as it prevented deployments in the past for my project as well and required manual intervention and reinstallation of rabbitMQ with data loss)