Hey everyone! I'm running into a problem while trying to create a new RabbitMQ cluster using Terraform. The issue seems to be related to the default namespace. When I attempt to deploy RabbitMQ there, it fails without generating any logs. Interestingly, when I use Helm to install it, everything works fine. I've used the same Terraform code for other clusters before without any issues. Does anyone have insights or solutions? Thanks!
3 Answers
Could you clarify what you mean by 'break'? A lot of failures with Kubernetes installations in Terraform happen because there aren't any nodes available, so the installation just fails silently. That's why I usually prefer using GitOps tools over deploying manifests through Terraform.
To get to the bottom of this, I need a bit more info:
1. Is your Kubernetes cluster stable and up when you're trying to deploy RabbitMQ?
2. What happens during the deployment? Does it fail before deployment, or is it just not getting to the Ready state?
I'd recommend sharing a link to your Infrastructure as Code files, any relevant logs, and some specific details. And just a tip, deploying to the default namespace isn't the best practice!
Yes, the cluster is up and all other pods are fine, except for RabbitMQ in the default namespace. The pods start creating but produce no logs. When I run it using Helm, it deploys smoothly. Thanks for the heads-up about the default namespace; I'll change that once I sort this out!
Using Terraform to install a Helm chart might be where your issues stem from. Have you tried troubleshooting from that angle?
The pods are stuck in creating mode and eventually crash without any logs in the default namespace. When I try another namespace, it works fine and I can access the logs. It's puzzling since other pods are running well.