Why does Azure reserve 5 IP addresses in every subnet?
Azure reserves the first four and the last IP address in every subnet. The first is the network address, the second is the default gateway (.1), the third and fourth are mapped to Azure DNS (.2 and .3), and the last is the broadcast address. These five cannot be assigned to any resource. A /29 with 8 total addresses only gives you 3 usable IPs — plan accordingly and avoid very small subnets where the 5-address overhead is significant.
What address space should I use for my Azure VNet?
Use a private RFC 1918 range — 10.0.0.0/8, 172.16.0.0/12, or 192.168.0.0/16. For most production VNets, starting with a /16 from the 10.x space gives 65,536 addresses and room to grow. If you plan to connect this VNet to other VNets via peering or to on-premises networks via VPN or ExpressRoute, ensure the ranges do not overlap — peered VNets cannot have overlapping address spaces.
What is the difference between the four subnet types in this calculator?
Gateway subnets host Azure VPN Gateways or ExpressRoute Gateways and must be named GatewaySubnet. Public subnets hold internet-facing services such as Azure Application Gateway or Azure Load Balancer with a public IP. Private subnets host virtual machines, App Services, and other workloads that only need outbound internet access through a NAT Gateway or firewall. Service subnets are delegated to specific Azure PaaS services — AKS node pools, Azure SQL Managed Instance, Azure Firewall — each with their own sizing and naming requirements from Microsoft.
Can I resize or delete a subnet in Azure after creating it?
You can only resize a subnet if it contains no resources. Once a VM, NIC, or service endpoint is deployed into a subnet, Azure will not allow you to change its address range. Deleting and recreating a subnet requires first removing all resources inside it. This is the most common and most painful Azure networking mistake — always plan your subnet sizes generously before deploying anything into them.
How does this calculator handle the VLSM allocation order?
The calculator sorts your subnets from largest required block to smallest before allocating addresses. This is the correct VLSM approach — placing the largest subnets first minimises boundary-alignment waste and ensures all subnets fit within the VNet address space without gaps. Your subnets are then re-sorted by network address for display, so the results table shows them in the natural sequential order they occupy in memory.
What is the Free / Unallocated IP Blocks table?
After all your subnets are placed, the remaining address space inside your VNet CIDR is decomposed into boundary-aligned CIDR blocks and shown in the free table. These are real, usable address ranges — not gaps or wasted space. You can assign any of these blocks to a new subnet at any time without disrupting existing subnets. The blocks are listed in sequential address order, smallest first, so the first entry is the one immediately after your last allocated subnet.
What is the smallest subnet size Azure allows?
The smallest subnet Azure supports is a /29, which has 8 total IP addresses. After the 5 Azure-reserved addresses, only 3 usable IPs remain. This is only suitable for very constrained infrastructure subnets — for example, a point-to-site VPN subnet or a small Private Endpoint subnet. Do not use /29 for any subnet that might need to scale. For most workloads, /27 (27 usable) is the practical minimum.