Opscraft
IP addressing & subnetting · Lesson 8 of 9 · 3 min read

Overlaps and real-world gotchas

Problem Effect Fix
Two VPCs both 10.0.0.0/16 Can't peer or route between them via TGW Re-plan with non-overlapping ranges; PrivateLink for specific services; NAT tricks as a last resort
Office uses 192.168.0.0/16, VPC uses 192.168.1.0/24 VPN routing breaks for some hosts Avoid common home/office ranges for cloud
Docker default bridge 172.17.0.0/16 clashes with a corporate range Hosts can't reach those corporate IPs Set bip and default-address-pools in /etc/docker/daemon.json
Kubernetes service CIDR overlaps a database network Pods can't reach the DB Choose service CIDR deliberately at cluster creation
Subnet exhausted during autoscaling New nodes/pods fail Bigger subnets, prefix delegation, secondary CIDRs, alerts on free IPs
Gateway outside the host's subnet No off-subnet connectivity Fix the gateway or mask
Enlarge diagramVPC A10.0.0.0/16VPC B10.0.0.0/16Re-IP one VPCpainful, but cleanPrivateLinkshare just one service via an NLBNAT trickslast resortpeering or TGW: rejected
Figure 1. Two VPCs with the same CIDR cannot be peered or routed together: a route to 10.0.x.x would be ambiguous.
Swipe sideways to see all of it, or tap to open it full screen.