Autoscaling workspace nodes not working even with scaler enabled on WS-Scheduler

I am using Gitpod self-hosted v0.8 on AWS. When I try to launch a workspace I get the following error(there are already a good number of active workspaces running in the deployment)

Oh, no! Something went wrong!

OutOfcpu: Pod Node didn't have enough resource: cpu, requested: 2000, used: 31510, capacity: 31850

From my understanding, the ws-scheduler is responsible for scaling the nodes in such cases. But in my cluster this apparently is not working(there might be something that I am missing here).

Can anyone help me out with this?

Thanks :smiley:

Hi @sri-vathsa, I’m sorry, I don’t know the answer either. Maybe someone from the community has one.

Please note, that we currently do not support AWS.

Hi @corneliusludmann ,
Are there supposed to be 2 components in WS-Scheduler i.e., Scheduler and Scaler for this functionality? I am looking at the current deployment template for WS-Scheduler in charts and it shows only schduler component.