rabbitmq-peer-discovery-k8s fails. I still need to manually stop, join_cluster and start every node

9/19/2019

Deployed the sample yaml files but doesn't work as expected. I still need to run a batch job on the cluster to manually stop, join_cluster and start for every pod. In addition, is there a way to configure vhosts, users and credentials using rabbitmq.conf file? Otherwise I have to do it in kerbernetes batch job after all the pods have booted up.

I have checked Failing K8s rabbitmq-peer-discovery-k8s clustering and k8s, RabbitMQ, and Peer Discovery. My cluster DNS works fine as it is able to resolve kubernetes to kubernetes.default.svc.cluster.local

-- Kok How Teh
kube-dns
kubernetes
kubernetes-statefulset
peer-discovery
rabbitmq

1 Answer

9/19/2019

I would suggest you install RabbitMQ using the proper Chart, even the prerequisite is having Helm installed.

-- prometherion
Source: StackOverflow