AKS: Support outbound IPs when using standard loadbalancer

[tombuildsstuff]

Re-basing #4400 - originally by @evenh

Fixes #4322

[katbyte]

LGTM 👍

[katbyte]

tests fail:

[evenh]

We need this functionality urgently. Is there anything we can do to help out @tombuildsstuff?

[Bowbaq]

I'd also really like this. Here's the workaround I have in the meantime:

data "azurerm_resource_group" "this" {
  ...
}

resource "azurerm_public_ip" "cluster" {
  ...
}

resource "azurerm_kubernetes_cluster" "this" {
  ...

  # Temporary hack until https://github.com/terraform-providers/terraform-provider-azurerm/issues/4322 is fixed
  provisioner "local-exec" {
    interpreter = ["bash", "-c"]
    command     = <<EOF
      if [ -z "$(az extension list | jq -r '.[] | select(.name == "aks-preview")')" ]; then
        az extension add -n aks-preview -y
      else
        az extension update -n aks-preview || true
      fi

      az aks update \
        --resource-group ${data.azurerm_resource_group.this.name} \
        --name ${self.name} \
        --load-balancer-outbound-ips ${azurerm_public_ip.cluster.id}
EOF
  }
}

[landro]

Hi @tombuildsstuff , any chance we can have this feature merged any time soon?

[landro]

Hi @tombuildsstuff - is there anything we can do to help move this PR forward?

[ekarlso]

@tombuildsstuff can we get this expedited for a release soon? We're having fun with duplicate ips on all our clusters that has Standard SKU.

[katbyte]

LGTM 👍

[seboss666]

Hello,
I've just ran into the problem of "duplicate IP" when switching to standard LB SKU (to activate API Server authorized IP ranges . Any update on conflict resolution ?

[ekarlso]

@tombuildsstuff any idea on this?

[landro]

Hi @tombuildsstuff !
this PR has been pushed from milestone to milestone for the past few months. Do you think it will make it this time? Is there anything we can do to help out?

[tombuildsstuff]

@landro unfortunately this isn't something we've got time to rebase at the moment - we're hoping that we can come back to this once the remaining bits of 2.0 have come together.

From our side if someone has time to rebase/fix this up/send a new PR we're more than happy to close this PR in favour of another one - however I don't believe we're going to get a chance to do this in the next couple of weeks unfortunately.

[lamdor]

just as an FYI, @evenh has rebased this into a new PR: #5394

[katbyte]

thanks @rubbish, closing in favour of #5394.

[ghost]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.

If you feel this issue should be reopened, we encourage creating a new issue linking back to this one for added context. If you feel I made an error 🤖 🙉 , please reach out to my human friends 👉 hashibot-feedback@hashicorp.com. Thanks!