# Asymmetric Routing

Asymmetric routing occurs when traffic traveling from a source to a destination follows a different network path from the return traffic.

Cloud concept. Updated 2026-10-29. Page: https://blowstack.com/tools/cloud-concepts-hub/asymmetric-routing

> Routing decisions are made independently in each direction, so a connection does not necessarily return through the devices that handled its outbound packets. This can be harmless in ordinary IP forwarding but can disrupt stateful firewalls, network address translation, and security inspection that depend on seeing both directions of a flow.

Practice exams covering Asymmetric Routing: [ANS-C01 AWS Certified Advanced Networking - Specialty](https://blowstack.com/practice-exams/aws-certified-advanced-networking-specialty), [SOA-C03 AWS Certified CloudOps Engineer - Associate](https://blowstack.com/practice-exams/aws-certified-cloudops-engineer-associate), [SCS-C03 AWS Certified Security - Specialty](https://blowstack.com/practice-exams/aws-certified-security-specialty), [SAP-C03 AWS Certified Solutions Architect - Professional](https://blowstack.com/practice-exams/aws-certified-solutions-architect-professional).

## 6 key aspects of Asymmetric Routing

Asymmetric routing becomes a security or availability problem when independently selected paths bypass required controls or separate packets from their connection state.

1. **Independent directional routing**: Forward and return packets are routed independently, using the routing information available at each hop. Inspect both directions rather than assuming that a successful outbound route guarantees a valid return path.
2. **Stateful inspection affinity**: Both directions normally need to reach the same stateful inspection instance or a system with appropriately shared connection state. Otherwise, valid responses can appear unsolicited or bypass required inspection.
3. **Routing and filtering are separate controls**: An allow rule cannot create a route, and a correct route cannot override a deny rule. Troubleshooting must distinguish path selection from security group, network ACL, and firewall policy evaluation.
4. **Translation state dependency**: Return packets must reach the network address translation state associated with the original connection. Sending responses through an unrelated translator can break a session even when the destination and security rules are correct.
5. **Multi-zone and failover behavior**: Load distribution, route convergence, and availability-zone selection can change which appliance receives a flow. Validate that normal operation and failover preserve the required inspection path, and account for possible resets of established sessions.
6. **Symmetry at the relevant processing points**: Not every intermediate router must be identical in both directions. The practical requirement is consistent traversal of devices whose inspection, translation, or policy enforcement depends on bidirectional visibility.

## Cloud services for Asymmetric Routing

Amazon VPC route tables establish the forward and return paths, while security groups and network ACLs determine whether traffic is permitted at their respective enforcement points. AWS Transit Gateway can connect workload networks to a centralized inspection VPC, with appliance mode supporting flow affinity across the inspection attachment. AWS Network Firewall provides managed inspection, while Gateway Load Balancer supports service insertion for virtual appliances. These designs still require deliberate bidirectional routing: neither stateful allow behavior nor appliance affinity repairs a bypass route. Use flow logs and firewall telemetry together to distinguish routing failures, filtering decisions, and missing connection state.

- **[AWS Network Firewall](https://blowstack.com/tools/cloud-services-hub/aws-network-firewall)**: Stateful inspection requires both directions of a flow to traverse the same firewall endpoint. Configure workload, firewall-subnet, and gateway routing as applicable so responses neither bypass inspection nor reach a different endpoint.
- **[Amazon VPC](https://blowstack.com/tools/cloud-services-hub/amazon-vpc)**: VPC route tables determine next hops independently for traffic in each direction. Security groups provide stateful filtering, network ACLs provide stateless subnet filtering, and VPC Flow Logs help investigate observed traffic without proving complete end-to-end delivery.
- **[AWS Transit Gateway](https://blowstack.com/tools/cloud-services-hub/aws-transit-gateway)**: Transit Gateway route tables can steer inter-network traffic through a centralized inspection VPC. Appliance mode on that VPC attachment maintains flow affinity to the same attachment network interface and Availability Zone, but correct forward and return routes remain necessary.

## Asymmetric Routing exam questions

Real practice exam questions on Asymmetric Routing, with the correct answer and its explanation.

### Question 1 (SOA-C03)

**Question:**

A company connects two workload VPCs to an AWS Transit Gateway. All traffic between the workload VPCs must pass through an inspection VPC that contains a stateful firewall appliance in each of two Availability Zones. The appliances do not share connection state, and the solution must retain inspection capacity in both Availability Zones.

Both workload attachments are associated with a transit gateway route table that routes inter-VPC traffic to the inspection attachment. The inspection attachment is associated with a separate route table that routes inspected traffic to the destination workload attachment. Within the inspection VPC, subnet routes direct traffic through the firewall in the Availability Zone selected by the transit gateway. Security groups and network ACLs permit the required traffic.

Connections between workloads in different Availability Zones frequently time out. Packet captures show that a connection's TCP SYN passes through one firewall, but its SYN-ACK reaches the other firewall and is dropped because no matching connection state exists.

Which change should a CloudOps engineer make to resolve the issue while maintaining the inspection requirements?

**Correct answer:** Enable appliance mode on the inspection VPC's transit gateway attachment.

**Explanation:** Appliance mode maintains the same Availability Zone selection on the inspection attachment for the lifetime of a flow, including return traffic. With the described per-AZ firewall routing, both directions traverse the same stateful appliance, preventing drops caused by missing connection state while retaining multi-AZ inspection capacity.

### Question 2 (SAP-C03)

**Question:**

A company has an Amazon VPC with CIDR block 10.40.0.0/16. An application tier uses subnet 10.40.10.0/24 in Availability Zone A, and a processing tier uses subnet 10.40.20.0/24 in Availability Zone B. Both tiers initiate TCP connections to each other. Their subnet route tables currently contain the VPC local route and a default route to a NAT gateway.

The company deploys AWS Network Firewall endpoints in dedicated firewall subnets in both Availability Zones. A new security requirement mandates stateful inspection of all traffic between the two tiers while preserving the original source IP addresses. Security groups and network ACLs already permit the required traffic by CIDR, and the firewall policy permits legitimate connections. The architecture must not depend on connection-state synchronization between the two firewall endpoints.

Which routing configuration meets these requirements?

**Correct answer:** Add a route for 10.40.20.0/24 in the application subnet route table that targets the Availability Zone A firewall endpoint. Add a route for 10.40.10.0/24 in the processing subnet route table that targets that same endpoint. Retain the VPC local route in the endpoint's firewall subnet route table.

**Explanation:** The destination-subnet routes are more specific than the VPC local route, so they steer inter-tier traffic through the firewall. Both directions use the same endpoint, providing the stateful engine with the complete connection. The firewall subnet's local route delivers inspected packets to the destination subnet. Network Firewall does not perform source NAT, so original source addresses are preserved.

### Question 3 (SCS-C03)

**Question:**

A company connects an application VPC to its data center through AWS Transit Gateway and two dynamically routed AWS Site-to-Site VPN connections. Each VPN connection terminates on a different on-premises stateful firewall. The firewalls do not synchronize connection state. Both VPN connections advertise identical prefixes with equal BGP attributes. VPN ECMP is enabled on the transit gateway, and the on-premises routers also use equal-cost routing for the VPC prefix.

Application connections intermittently fail. Packet captures show that a TCP SYN leaves the data center through one firewall, but the SYN-ACK arrives through the other firewall and is dropped because no matching connection state exists. VPC security groups and network ACLs permit the required traffic.

The company must make new connections work reliably during normal operation while retaining automatic failover if the preferred VPN connection fails. Existing connections may be reestablished after failover. The company cannot replace the firewalls or enable state synchronization.

Which configuration meets these requirements?

**Correct answer:** Disable VPN ECMP on the transit gateway. Prepend the AS path for data center prefixes advertised through the backup VPN connection, and configure higher on-premises local preference for VPC routes received through the preferred VPN connection.

**Explanation:** The longer AS path makes AWS prefer the primary VPN connection for data center destinations, while on-premises local preference selects that same connection for VPC destinations. Both directions therefore traverse the same stateful firewall. BGP withdrawal allows the backup connection to take over after a failure, and the requirement permits connections to be reestablished then.

### Question 4 (ANS-C01)

**Question:**

A company connects VPC-A (10.40.0.0/16), VPC-B (10.50.0.0/16), and an inspection VPC to an AWS Transit Gateway. Company policy requires both directions of every connection between the application VPCs to traverse stateful appliances in the inspection VPC.

The transit gateway configuration is as follows:
• The VPC-A attachment is associated with RT-A, which contains only a 0.0.0.0/0 route to the inspection VPC attachment.
• The VPC-B attachment is associated with RT-B, which contains a 0.0.0.0/0 route to the inspection VPC attachment and a propagated 10.40.0.0/16 route to the VPC-A attachment.
• The inspection VPC attachment is associated with RT-Inspection, which contains propagated routes to both application VPC attachments.

Appliance mode is enabled on the inspection VPC attachment. All VPC subnet routes and appliance forwarding configurations have been verified as correct. Packet captures show that requests from VPC-A to VPC-B traverse an inspection appliance, but responses reach VPC-A without traversing an appliance.

Which change will meet the inspection requirement while preserving connectivity?

**Correct answer:** Disable propagation from the VPC-A attachment into RT-B and remove its propagated route from RT-B. Retain VPC-A propagation into RT-Inspection.

**Explanation:** Without the direct route to VPC-A in RT-B, responses entering through the VPC-B attachment follow RT-B's default route to inspection. After inspection, RT-Inspection provides the route to VPC-A. Requests already follow the equivalent inspection path through RT-A, so both directions traverse inspection while connectivity is preserved.

## Practice Asymmetric Routing in a full exam

Questions on Asymmetric Routing in each practice exam, with every answer option and an explanation for each:

- [ANS-C01](https://blowstack.com/practice-exams/aws-certified-advanced-networking-specialty): 6 questions
- [SCS-C03](https://blowstack.com/practice-exams/aws-certified-security-specialty): 4 questions
- [SOA-C03](https://blowstack.com/practice-exams/aws-certified-cloudops-engineer-associate): 1 question
- [SAP-C03](https://blowstack.com/practice-exams/aws-certified-solutions-architect-professional): 1 question

## What is Asymmetric Routing

Asymmetric routing describes a difference between the forward and return paths of a network conversation. For example, a request from an application to a remote server might pass through firewall A, while the response passes through firewall B. IP routing does not inherently require identical paths: each router selects a next hop using its own routing information and the packet's destination. The important security question is whether both directions traverse the inspection and translation state required for the connection.

Stateful devices track connections rather than evaluating every packet independently. A firewall that observes a TCP SYN can establish state for the subsequent handshake and data exchange. If the SYN-ACK reaches another firewall that does not share that state, the second firewall may drop it as an unexpected packet. Alternatively, return traffic might bypass inspection entirely. Asymmetry therefore can cause either availability failures or gaps in security enforcement, depending on the topology and policies.

In cloud networks, independent subnet route tables, centralized inspection networks, multiple availability zones, VPN connections, and dynamically learned routes can create asymmetric paths. Consider two workload networks connected through a transit hub. Traffic from the first network is routed through an inspection attachment, but the second network has a more specific route directly back to the first. The request reaches the destination through the firewall, while the response bypasses it. Adding an allow rule does not correct the underlying routing problem; both directions must be steered through the required inspection path.

With AWS Network Firewall, design routes so both directions of a flow traverse the same firewall endpoint. Multi-zone deployments require deliberate endpoint selection in the relevant route tables. For centralized inspection using AWS Transit Gateway, appliance mode on the inspection VPC attachment helps maintain flow affinity to the same attachment network interface and Availability Zone for the flow's lifetime. Appliance mode does not insert a firewall automatically or correct route tables that bypass inspection.

Security groups and network ACLs must be evaluated separately from path symmetry. Security groups are stateful and generally allow response traffic for tracked, permitted connections without a separate reverse-direction allow rule. That behavior does not create a missing return route or ensure traversal of a separate firewall. Network ACLs are stateless, so each direction needs explicit permission. For an HTTPS connection, this commonly means allowing the destination service port in one direction and the client's ephemeral destination port in the return direction, using the applicable client port range.

Network address translation introduces another dependency: return traffic must reach the translator that holds the mapping, or a system that has equivalent shared state. A route change that sends responses to a different translator can break existing connections even when firewall rules permit the traffic. Symmetry requirements therefore concern the relevant stateful processing points, not necessarily every router or physical link along the path.

Troubleshoot by identifying the connection's source and destination addresses, ports, and protocol, including any translated addresses. Trace forward and return routing decisions independently, checking longest-prefix matches, subnet associations, transit routing, and on-premises advertisements. Correlate flow logs, firewall logs, and packet captures where available; an accepted flow-log record at one interface does not prove end-to-end delivery. Test new and established connections after failover or route changes. Symmetric inspection can add routing complexity, latency, and cross-zone costs, but widening security rules is not a reliable substitute for correcting the path.

## Learn Asymmetric Routing in lessons

3 short lessons on Asymmetric Routing in 3 certification courses.

- SOA-C03: Asymmetric Routing: paths through stateful devices (Premium), https://app.blowstack.com/lessons/asymmetric-routing-paths-through-stateful-devices-soa-c03
- SAP-C03: Asymmetric Routing: preserving stateful inspection paths (Premium), https://app.blowstack.com/lessons/asymmetric-routing-preserving-stateful-inspection-paths-sap-c03
- SCS-C03: Asymmetric Routing: flow affinity through stateful inspection (Premium), https://app.blowstack.com/lessons/asymmetric-routing-flow-affinity-through-stateful-inspection-scs-c03
