Install Confluence on AWS via Kubernetes

[AM]

Hi all!

I am trying to install Confluence using  this documentation on AWS. I managed to complete all of the steps, but when Kubernetes at the end tries to schedule a confluence pod to run on one of the nodes I get an error (via command kubectl describe nodes) like

0/2 nodes are available: 2 Insufficient cpu. preemption: 0/2 nodes are available: 2 No preemption victims found for incoming pod.

I tried googling and supposedly this means that the pods have no available CPU capacity, so I event tried propping up 4 nodes, but still the same error.

There are definitely available resources so I'm confused by the message.

[AM · Accepted solution]

I finally managed to deploy this. For anyone stumbling over this I have a writeup on my wiki here https://dw.exitstatus0.com/doku.php?id=wiki:confluence_aws_deploy. Hope it helps someone else.

[Yevhen]

@AM perhaps your nodes do not have enough cpu or memory. You may want to override the default requests

Adding more nodes won't help if they are the same nodes with less than 2 vCPU available. You may need to use larger nodes or decrease cpu request.

[AM]

Hi Yevhen, thank you!

I will try that, but IMHO this deployment should have been tested by Atlassian, since these are the default values.

[Yevhen]

The deployments are continuously tested. The default values are close to prod values, and having 2 vCPU available isn't a huge resources request, imho.