Tero Marttila

Best workaround might be to use separate resources for the two structural cases: ```hcl resource "aws_lb_listener_rule" "http-forward" { count = (var.second_target_group_arn == null) ? 1 : 0 listener_arn = var.alb_listener_arn...

Is the fix for this issue shipped in any released zfs 2.1.* version? I am unable to find any reference to this issue or the fixing PR in the changelogs...

> On Hacker News, terom wrote: Disclaimer: this is just speculation on my part based on the changes in the patch... I may have misunderstood something, I didn't verify any...

There's some haphazard `namespace` support at the `Client` level, but it's not really very well tested. I'm not entirely sure how the `APIClient` should behave in terms of non-namespaced resources...

> IMO it should take that into account. In what kind of usages, as examples?

So that would be: ```ruby client = K8s::Client.config(K8s::Config.load_file(ENV['KUBECONFIG'])) client.create_resource(K8s::Resource.from_file('pod.yml')) ``` You can already get pretty close with `client.client_for_resource(resource, namespace: config.context.namespace).create_resource(resource)`, except it seems like: * `K8s::Client.config` doesn't set `@namespace` from...

This isn't as simple as just having `K8s::Client#client_for_resource` pass the client's default `@namespace` to the `ResourceClient`, because not all resources are namespaced... fixing this to work correctly for all methods...

There are three cases to consider: * `namespace: 'default'` for `GET/PUT/POST /api/v1/namespaces/default/services/foo` of namespaced resources * `namespace: nil` for `GET /api/v1/services` to list all namespaced resources * `namespace: nil` for...

Somewhat mitigated by the `/*/log/journal` being optional... if neither is mounted to the container, then the systemd execs will simply not read anything from the journal. I suppose we could...

Also questioning the rationale behind this decision to deprecate `-` in group names. Not being able to use dashes in dynamic inventory `keyed_groups` was already annoying enough, but having to...