We’re using Fly.io for automatic feature branch deployments based on Github PRs, but we currently have to manually scale the VM each time as it runs out of memory.
Is there a better way to do this? Can we specify the required VM size on deploy/launch?
We’re actually using a custom action made in an Earthfile, but inspired by that action.
I see in that PR that they are also just running a separate fly scale. We’re doing the same, but still having issue. Looks like this is the only way to do this at the moment. Thanks!