Hi @Fabio ,
this answer from Romaric contains all the info:
- Your cluster has 2CPU / 4GB RAM but part of it is already used by the installed qovery applications (cert manager, loki etc…).
- when deploying a new version of the application, a new instance is created so you’ll have temporary twice the number of CPU and ram allocated to your app.
so the solution here is to increase the node size or to change the deployment strategy of your app to Recreate
(which means that your service won’t be available during the rollout of a new version)