Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Compute Instance with no Public IP fails to provision #85

Open
RoozbehBandpey opened this issue Mar 8, 2022 · 1 comment
Open

Compute Instance with no Public IP fails to provision #85

RoozbehBandpey opened this issue Mar 8, 2022 · 1 comment

Comments

@RoozbehBandpey
Copy link

RoozbehBandpey commented Mar 8, 2022

Context (Environment)

I'm trying to manually create a compute instance with no public IP address (since it's not possible to do it with Terraform). After setting up a private Azure ML workspace following this template 201-machine-learning-moderately-secure

Terraform (and AzureRM Provider) Version

terraform v0.14.0
azurerm v2.96.0

Current Behavior

Provisioning fails with this error:

Provisioning error
The specified subnet /subscriptions/[REDACTED]/resourceGroups/[REDACTED]/providers/Microsoft.Network/virtualNetworks/[REDACTED]/subnets/snet-training-dev-westeurope has PrivateLinkServiceNetworkPolicies or PrivateEndpointNetworkPolicies enabled. Please disable them to provision cluster/instance with no public IP. Please read this document for more details: https://aka.ms/AMLPLNetPolicies

Provisioning Compute Instance with public IP address completes successfully.

Possible Solution

I have tried every combination of PrivateLinkServiceNetworkPolicies or PrivateEndpointNetworkPolicies on the training subnet. No luck!

Here's my training Subnet block

resource "azurerm_subnet" "snet_training" {
  name                                            = "snet-training-${var.deploy_target}-${var.machine_learning_location}"
  resource_group_name                             = data.azurerm_virtual_network.existing_mlw_vnet.resource_group_name
  virtual_network_name                            = data.azurerm_virtual_network.existing_mlw_vnet.name
  address_prefixes                                = var.snet_training_addr_prefixes
  enforce_private_link_endpoint_network_policies  = true 
  enforce_private_link_service_network_policies   = false 
}

Steps to Reproduce

  1. 201-machine-learning-moderately-secure
@vrugelda
Copy link

Did you solve this issue?

I'm deploying the same as you and hitting a couple of errors. Most likely network related.
As of now I'm not deploying with this No Public IP option..
When I'm trying to connect to a notebook from an instance (logged on to the Jumpbox) - it complains about Kernel error and network. (even access as well)

Could this be related to the same policies.. ?

Any tip would be helpful

error

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants