Ignore AMI updates on etcd/master nodes to prevent destroying nodes accidently

[chancez]

Currently if a new version of CoreOS comes out on a user's configured
channel after they've already deployed, and they re-run terraform apply,
Terraform will detect the new AMI, and when reconciling desired state,
it will attempt to destroy existing nodes to update the AMI.

In order to avoid this, I added a ignore_changes to both the etcd node
resource, and the master launch configuration resource, to avoid
updating masters/etcd nodes if the AMI changes.

Longer term, I would expect the Container Linux Update Operator, or a
terraform operator could resolve this, but this seems like a reasonable
change to prevent accidental destruction of a cluster.

Previously this change also set prevent_destroy, but that cannot be
parameterized with vars, so it was removed.

[chancez]

I was going to make changes to parameterize the force_destroy, but ran into issues and found this:

hashicorp/terraform#10730

It seems there's no way to easily prevent destruction of a resource during apply, but allow for it during destroy today. I've updated my PR remove the force_destroy bit, as the ignore_changes should handle the case I was seeing, but it would be good if we could find a way to prevent accidental deletion of etcd nodes on a terraform apply, as I've done so a few times now by accident, even after changing nothing in my vars. Perhaps we can contribute something to upstream similar to what's discussed in hashicorp/terraform#3116.

[s-urbaniak]

LGTM, good catch, but I would like to have @alexsomesan have another look.

[alexsomesan]

I agree with @s-urbaniak. It's a good fix.
Users aren't left out in the dark past anyway if they keep CL updates enabled.
Let keep doing this until we have a more solid story around it based on Container Linux Update Operator or the like.

LGTM.

[sym3tri]

Why not add to workers as well?

[s-urbaniak]

@sym3tri good catch! @chancez do you mind to submit another PR which includes workers?

[chancez]

I figured I'd keep the scope small since it's a no brainer for etcd and masters, but wasn't sure about workers. I'd be glad to make another PR with the same change.