Skip to content

Lesson 08 — East-West Traffic Protection

By the end of this lesson, you will be able to:

  • Understand East-West traffic in Kubernetes
  • Differentiate East-West and North-South traffic
  • Understand why East-West traffic is a major attack vector
  • Learn how attackers perform lateral movement
  • Secure East-West traffic using Network Policies
  • Protect service-to-service communication using Service Mesh and Mutual TLS (mTLS)
  • Design enterprise East-West security architectures
  • Apply enterprise best practices

Modern applications are built using hundreds or even thousands of microservices.

While organizations spend significant effort protecting Internet-facing applications, most cyber attacks occur after an attacker gains an initial foothold.

Once inside a Kubernetes cluster, attackers attempt to move laterally by accessing:

  • APIs
  • Databases
  • Authentication services
  • Secrets
  • Internal applications
  • Monitoring systems

This internal communication is called East-West traffic.

Without proper controls, attackers can move from one compromised workload to another until they gain access to critical business systems.

Protecting East-West traffic is a fundamental component of Zero Trust security.


East-West traffic refers to communication between workloads inside the same environment.

Examples include:

  • Pod to Pod
  • Service to Service
  • Namespace to Namespace
  • Node to Node
  • Cluster to Cluster

Example:

Frontend Pod
API Pod
Payment Service
Database

All of this communication occurs internally.


North-South traffic refers to communication entering or leaving the Kubernetes cluster.

Example:

Internet
Application Load Balancer
Ingress Controller
Frontend Application

North-South traffic connects users with applications.


East-West North-South
Internal communication External communication
Pod-to-Pod Internet-to-Cluster
Service-to-Service Client-to-Application
Higher attack frequency Easier to monitor
Lateral movement Initial access

Both traffic flows require strong security controls.


After compromising a single Pod, attackers rarely stop there.

Instead they attempt to:

  • Discover additional services
  • Locate databases
  • Find Secrets
  • Access internal APIs
  • Escalate privileges
  • Reach Kubernetes control components

Example:

Compromised Pod
API
Database
Secrets
Critical Applications

Without segmentation, one compromised workload can lead to complete cluster compromise.


Lateral movement occurs when an attacker moves from one system to another after initial compromise.

Example:

Web Application
API
Order Service
Payment Service
Database

Every successful step increases the impact of the attack.


Zero Trust assumes:

Internal traffic should never be trusted automatically.

Instead of:

Internal Traffic
Trusted

Use:

Request
Authenticate
Authorise
Encrypt
Allow

Every request is verified regardless of its source.


Enterprise Kubernetes environments secure East-West traffic using multiple layers.

Network Policies
Service Mesh
Mutual TLS
Authorization Policies
RBAC
IAM
Continuous Monitoring

Each layer strengthens security and reduces the attack surface.


Network Policies provide the first layer of East-West protection.

Example:

Frontend
API
✓ Allowed
-------------------
Frontend
Database
✗ Denied

Only approved communication paths are permitted.


Mutual TLS ensures every service verifies the identity of the service it communicates with.

Example:

Service A
Certificate Verification
Encrypted Connection
Certificate Verification
Service B

If identity verification fails, communication is denied.


A Service Mesh such as Istio or Linkerd secures East-West communication.

Capabilities include:

  • Automatic mTLS
  • Identity verification
  • Authorization Policies
  • Traffic encryption
  • Service authentication
  • Traffic observability
  • Policy enforcement

The Service Mesh protects internal application communication without requiring application code changes.


Traditional networks trust IP addresses.

Cloud-native environments trust identities.

Instead of:

10.0.5.18

Service Mesh verifies:

payment-service
Verified Identity
order-service

Because Pods are frequently recreated, identities are far more reliable than IP addresses.


Namespaces provide logical security boundaries.

Example:

Finance Namespace
Finance Database
✓ Allowed
--------------------
HR Namespace
Finance Database
✗ Denallowed

Namespaces reduce the blast radius of security incidents.


Micro-segmentation divides applications into secure communication zones.

Example:

Frontend
API
Payment
Database

Blocked communication:

Frontend
Database
✗ Denied

Only required communication paths are allowed.


Frontend
Network Policy
Envoy Sidecar
Mutual TLS
API
Envoy Sidecar
Database

Every internal request is:

  • Authenticated
  • Encrypted
  • Authorised
  • Logged

Amazon EKS
Namespaces
Network Policies
Istio
Envoy Sidecars
Mutual TLS
Application Pods

Amazon EKS combines Kubernetes networking with Service Mesh technologies to secure internal communications.


A multinational bank operates over 600 microservices within Amazon EKS.

Security architecture includes:

  • Network Policies
  • Istio
  • Automatic Mutual TLS
  • Namespace isolation
  • Authorization Policies
  • Continuous monitoring
  • CloudWatch integration

Communication flow:

Customer Portal
API Gateway
Payment Service
Fraud Detection
Customer Database

Every service verifies the identity of the next service before communication is allowed.

This architecture supports Zero Trust principles and regulatory compliance.


Cloud Security Engineers frequently identify:

  • Missing Network Policies
  • Disabled Mutual TLS
  • Overly permissive communication
  • Flat network architecture
  • Missing Namespace isolation
  • Unencrypted service communication
  • Excessive service permissions
  • Poor Service Mesh configuration
  • Lack of traffic monitoring
  • Identity spoofing

These weaknesses allow attackers to move laterally through the environment.


Security teams should monitor:

  • East-West traffic volume
  • Failed Mutual TLS authentication
  • Authorization Policy violations
  • Unexpected service communication
  • Cross-Namespace traffic
  • Lateral movement indicators
  • Certificate failures
  • Sidecar health
  • Network Policy violations
  • Service identity changes

Telemetry from Istio, Linkerd, Envoy, Amazon CloudWatch, Prometheus and Grafana provides deep visibility into internal communications.


A recommended enterprise approach:

Step 1
Segment Applications
Step 2
Deploy Network Policies
Step 3
Implement Service Mesh
Step 4
Enable Mutual TLS
Step 5
Apply Authorization Policies
Step 6
Monitor Internal Traffic
Step 7
Review Communication Paths Regularly

This layered approach significantly reduces the risk of lateral movement.


As a Kubernetes Security Engineer:

  • Implement a Zero Trust networking model.
  • Enable Mutual TLS for all service-to-service communication.
  • Use Network Policies to enforce least-privilege networking.
  • Isolate workloads using Namespaces.
  • Implement micro-segmentation between applications.
  • Authenticate workloads using Service Mesh identities.
  • Continuously monitor East-West traffic.
  • Audit communication paths regularly.
  • Investigate unexpected service communication immediately.
  • Keep Service Mesh components updated.

Strong East-West protection limits the impact of compromised workloads and improves the overall security posture of Kubernetes clusters.


A vulnerable inventory application in an Amazon EKS cluster is compromised through a remote code execution vulnerability.

The attacker attempts to:

  • Access the payment service
  • Reach customer databases
  • Connect to the Kubernetes API
  • Discover additional workloads

However, the organisation has implemented:

  • Network Policies
  • Namespace isolation
  • Istio Service Mesh
  • Automatic Mutual TLS
  • Authorization Policies
  • Continuous traffic monitoring

Every unauthorised request is blocked because:

  • The compromised workload lacks permission.
  • Service identities cannot be spoofed.
  • Communication is encrypted and authenticated.
  • Security teams receive immediate alerts through monitoring dashboards.

The attacker is unable to move laterally, and the incident is contained to a single compromised Pod.


After completing this lesson, you should understand:

  • What East-West traffic is
  • How it differs from North-South traffic
  • Why attackers target internal communication
  • The importance of preventing lateral movement
  • How Network Policies protect East-West traffic
  • How Service Mesh and Mutual TLS strengthen internal security
  • The role of micro-segmentation and Namespace isolation
  • Enterprise monitoring and best practices

East-West traffic protection is a cornerstone of Kubernetes security. By combining Network Policies, Service Mesh, Mutual TLS, identity-based networking and continuous monitoring, organisations can implement Zero Trust networking and dramatically reduce the risk of lateral movement within Amazon EKS environments.


What is East-West traffic in Kubernetes?

  • A. Traffic entering the cluster from the Internet
  • B. Traffic between workloads inside the Kubernetes environment
  • C. Traffic between AWS Regions
  • D. Traffic between Availability Zones only

Answer: B


Which attack technique is most closely associated with East-West traffic?

  • A. SQL Injection
  • B. Lateral Movement
  • C. DNS Amplification
  • D. Cross-Site Scripting (XSS)

Answer: B


Which Kubernetes feature is commonly used to restrict Pod-to-Pod communication?

  • A. ReplicaSets
  • B. Network Policies
  • C. Persistent Volumes
  • D. Storage Classes

Answer: B


Which technology automatically encrypts and authenticates service-to-service communication?

  • A. kube-proxy
  • B. Mutual TLS (mTLS)
  • C. CoreDNS
  • D. Amazon Route 53

Answer: B


Which of the following is considered an enterprise best practice for protecting East-West traffic?

  • A. Allow unrestricted communication between all workloads.
  • B. Trust all internal traffic by default.
  • C. Implement Network Policies, Service Mesh, Mutual TLS and continuous monitoring.
  • D. Disable Namespace isolation to simplify networking.

Answer: C


In the next lesson, you will learn about Zero Trust Networking, exploring how enterprise organisations apply Zero Trust principles across Amazon EKS by combining workload identity, continuous verification, least-privilege access, Service Mesh, Network Policies, IAM, RBAC and policy enforcement to build resilient cloud-native environments.

➡️ Next Lesson: Lesson 09 — Zero Trust Networking