SOLVED

What determines the outbound ip addresses in my app service in Azure?

Copper Contributor

I have one app service in Azure in a particular region under Premiumv2 Central US and a second in North Central US  region under Premium V3. 

If i navigate to the Networking in the first app in Azure portal under Networking ->OutBound addresses-> Show More, there are about 8 outbound Ip addresses.

If i navigate to the Networking in the second app in Azure portal under Networking ->OutBound addresses-> Show More, there are about about 20 outbound Ip addresses.

 

The second app service is a replica of the first , so same .net code. The first app has been scaled but the second has not 

Why is there such a big difference in the number of outbound ip addresses. 

 

What determines what ip address is populated in the the outbound ip addresses ?

1 Reply
best response confirmed by KarenSaund888 (Copper Contributor)
Solution
Each Region and service has a designated IP range: https://www.microsoft.com/en-us/download/details.aspx?id=56519

"An App Service app runs in an App Service plan, and App Service plans are deployed into one of the deployment units in the Azure infrastructure (internally called a webspace). Each deployment unit is assigned a set of virtual IP addresses, which includes one public inbound IP address and a set of outbound IP addresses. All App Service plans in the same deployment unit, and app instances that run in them, share the same set of virtual IP addresses. For an App Service Environment (an App Service plan in Isolated tier), the App Service plan is the deployment unit itself, so the virtual IP addresses are dedicated to it as a result."

I can't see any reference that designates you will have 20 IPs for V3, but I believe the higher scale sizing allows additional scale-out capability so you get assigned those IP for redundancy and throughput.
https://docs.microsoft.com/en-us/azure/app-service/overview-inbound-outbound-ips
1 best response

Accepted Solutions
best response confirmed by KarenSaund888 (Copper Contributor)
Solution
Each Region and service has a designated IP range: https://www.microsoft.com/en-us/download/details.aspx?id=56519

"An App Service app runs in an App Service plan, and App Service plans are deployed into one of the deployment units in the Azure infrastructure (internally called a webspace). Each deployment unit is assigned a set of virtual IP addresses, which includes one public inbound IP address and a set of outbound IP addresses. All App Service plans in the same deployment unit, and app instances that run in them, share the same set of virtual IP addresses. For an App Service Environment (an App Service plan in Isolated tier), the App Service plan is the deployment unit itself, so the virtual IP addresses are dedicated to it as a result."

I can't see any reference that designates you will have 20 IPs for V3, but I believe the higher scale sizing allows additional scale-out capability so you get assigned those IP for redundancy and throughput.
https://docs.microsoft.com/en-us/azure/app-service/overview-inbound-outbound-ips

View solution in original post