“So to complete our configuration we just need your outbound static IP…” This is something that pops up again and again, specially if you work integrating legacy systems, like banks, government agencies or other older systems that requires a static IP Address to add to firewall inbound rules.
In the past I had to use on of the subscription tiers from Azure API Management or in some cases deploy the code within a self hosted service in a Virtual Machine. And both of them are valid options if you already have one of those components in place. But in a couple of my last projects, Not only I needed to implement a new component (in this case API Management) just to fulfil this requirement. I also had to change my design a bit, because API Management didn’t support one of the streaming requirements I had (but that is a story for another post).
But I wouldn’t be writing a post just to complain about this (not that this has never happened before) – I actually found another solution – one that I’ve tried and discarded earlier this year, but thanks to some good work from the App Services Engineering team, is finally a viable solution: Azure Functions + VNET Integration + NAT Gateways!
How does it work?
According to Microsoft Docs:
“NAT gateway resources are part of Virtual Network NAT and provide outbound Internet connectivity for one or more subnets of a virtual network. The subnet of the virtual network states which NAT gateway will be used. NAT provides source network address translation (SNAT) for a subnet. NAT gateway resources specify which static IP addresses virtual machines use when creating outbound flows.”Designing virtual networks with NAT gateway resources
So, the NAT Gateways provides a way to specify a static IP addresses to be used as outbound address to any outbound traffic that is leaving a subnet towards the internet. Sounds like something we would like to do right?
When I first learned about this resource around the end of 2019, I’ve tried to configure the following scenario:
But I got stuck on an issue – NAT Gateway were not compatible with the App Service Plans for either Azure Function Consumption Premium or Dedicated plans. Doing some research I found that the NAT Gateway at that point was not compatible with the basic load balancer and any products that used basic load balancer.
But just this week, the App Services team provided us with some good news:
You can read their blog post here. So whatever incompatibility between the NAT Gateway and the App service plans are now solved! So let’s try to get it working…
In order to implement this you will need the following items:
- An Azure Function configured with an App Service plan that supports VNET integration (Premium Consumption, or one of the following dedicated plans: Standard, Premium V2 or Premium V3),
- A Virtual Network with at least one available subnet,
- A standard IP Address and
- A NAT Gateway.
You can find more information about how to configure the NAT Gateway and associate it to a subnet here. The configuration is pretty simple – just requires the standard IP address and the VNET/Subnet associated.
Notice that a single NAT Gateway can be connected to multiple subnets. So as long as applications requires the same outbound IP addresses, they can be integrated to subnets connected to the same NAT Gateway.
Azure Function Configuration
The configuration for Azure Functions is quite straightforward. First you need to go to Networking (1) and select configuration (2).
After that you need to click on + Add Vnet (1), then select an existing Virtual Network (2), click on select existing (3) and choose one of the available subnets (4) and finally click Ok (5).
Once the VNET configuration is completed, it should look like this:
To confirm that the subnet is configured to connect to the NAT Gateway, you can click on the subnet name, which will bring you to the subnet configuration page. Here, you can confirm that the NAT Gateway is assigned to that subnet.
Testing your configuration
A simple way to test your configuration is to make a http request to a website that would return the IP address of the caller. I’ve used a PowerShell script to do that, taking the advantage of Azure Function support for PowerShell Core. You can find the code in the gist below:
A whole new world…
The ability to have Azure Functions and Web Apps having a static outbound IP is a feature that have been requested for some time now. This will open up a number of scenarios that were not possible today with pure serverless and will simplify a lot the integration with other systems, in special legacy and on-premises systems that still depend on a static IP address to configure inbound access.
That alone would already be fantastic news! But another recent announcement from the Enterprise Integration team combined with this makes those news even more exciting… The new Logic Apps runtime, which is an extension of Azure Functions Runtime, can also leverage from this configuration. So not only core Azure Functions implementations can take advantage of this implementation, but Logic Apps will soon be able to leverage from that as well. Actually, I’ve tested the same configuration on a logic apps (preview) resource and it behave as expected!
I’ve been doing a bit more research on the Logic Apps (preview) and what I found so far is that the managed connectors are still using their own IP addresses. For example, I’ve tried to connect to the an FTP server using the SFTP-SSH connector and it presented an IP from the managed connector in Australia East region. Hopefully this is something that will be addressed as the preview moves close to GA. Otherwise we will have to replace connector functionality with Azure Functions (still better than the original situation, but not ideal).
- App Services can now leverage from the NAT Gateway functionality to expose a static outbound IP address to the internet.
- Azure Function with either a Premium Consumption or Dedicated app service plan can also leverage from the same functionality.
- Logic Apps (preview), which are based on the Azure Functions runtime will also benefit from the same type of configuration.
- This will open up a number of scenarios that were not easily implemented before – in special when integrating with on-premises or legacy systems.