I'm not sure if this is possible but basically when

ElasticBeanstalk

creates an

Environment

, one of the outputs is

autoscalingGroups

. I'd like to update the auto scaling group(s) to enable metrics collection. This doesn't appear to be possible in the EBS initialisation options so I think i might need to update it afterwards. i tried something like this:

const autoScalingGroup = Group.get(`autoscaling`, environment.autoscalingGroups[0]);
    autoScalingGroup.enabledMetrics = [`GroupDesiredCapacity`, `GroupInServiceCapacity`, `GroupPendingCapacity`, `GroupMinSize`, `GroupMaxSize`, `GroupInServiceInstances`, `GroupPendingInstances`, `GroupStandbyInstances`, `GroupStandbyCapacity`, `GroupTerminatingCapacity`, `GroupTerminatingInstances`, `GroupTotalCapacity`, `GroupTotalInstances`];

but the property is readonly

It isn't possible to update any resource in Pulumi except via its constructor, and this automatically-generated ASG is not managed by Pulumi. You need to use the SDK for this.

ok.. so would the general process be to do something like: • add the autoscaling group as an output from the stack • after running

pulumi up

, run (something like) `aws autoscaling enable-metrics-collection --auto-scaling-group-name `pulumi stack output autoscalingGroup`` and that would form part of the 'build' script?

Yes. Or you could also wrap this in the automation API.

could you elaborate on that pls Itay?

Sorry, didn’t mean to be overly terse. 🙂 The benefit of the Pulumi automation API is that you can invoke your stack programmatically, reference the output programmatically, and then do something with the output. So in your case, you could create your stack by invoking the API, and then get the output of the stack (the ASG), and then call the AWS SDK with that ASG. An example that we do is we send an email to a user based on the stack output.

no need to apologise 🙂 thanks heaps for the clarification though!