Kubernetes East West Traffic
Kubernetes east west traffic is network flow that moves between pods inside your cluster. You control this flow by setting up network rules and strict role permissions. By default, every pod in a cluster can talk to every other pod. Teams often start with this open setup because it makes early app testing fast. Still, open internal networks let a single bad pod reach every other service. Broad cluster roles also let users jump between workloads during live debugging. How kubernetes east west
Your coding agents run with your access.
Claude Code reaches prod with a developer's credentials, and approval prompts get clicked through. hoop.dev checks every query, masks sensitive data, and holds risky changes for a human.
Kubernetes east west traffic is network flow that moves between pods inside your cluster. You control this flow by setting up network rules and strict role permissions.
By default, every pod in a cluster can talk to every other pod. Teams often start with this open setup because it makes early app testing fast. Still, open internal networks let a single bad pod reach every other service. Broad cluster roles also let users jump between workloads during live debugging.
How kubernetes east west traffic works inside a cluster
North south traffic enters or leaves the cluster from the outside world. In contrast, kubernetes east west traffic stays inside the cluster boundary. When a frontend pod calls a backend pod, that request travels east to west. The cluster network routes packets from one pod IP address to another pod IP. This happens even when the two pods run on different cluster nodes.
Network rules are the primary native way to restrict pod to pod calls. Without network rules, all pods accept packets from any internal source. When you apply a network rule to a pod, it drops unlisted traffic. You can write rules that only allow traffic from specific namespaces or pod labels.
Human and service identity permissions also affect internal cluster movement. Kubernetes Role Based Access Control, or RBAC, governs who can call the cluster API. In RBAC, a Role sets permissions inside a single namespace. A ClusterRole sets permissions across all namespaces or for cluster-wide resources like nodes. A RoleBinding grants permissions inside one namespace, while a ClusterRoleBinding grants them cluster-wide.
Where native network rules and RBAC roles fall short
Native network rules and RBAC solve the first layer of internal security. Network rules block unwanted packets at the IP and port level. RBAC rules stop unauthorized users or pods from calling the cluster API server. Together, they define which pods can talk and who can reach the API.
Still, RBAC permissions are purely additive, meaning there are no deny rules. If a user gets a broad ClusterRoleBinding, you cannot add a small deny rule. You must rebuild their bindings from scratch to remove excess permissions.
East-West Traffic Security + Kubernetes RBAC: architecture patterns and best practices
Continue reading? Get the full guide.
Free. No spam. Unsubscribe anytime.
Subresources create another operational gap during live debugging and internal movement. RBAC represents subresources like pod logs or pod execution with a slash in the name. When an engineer has pod execution rights, they can open a shell inside a pod. Once inside that pod, the engineer inherits the pod's east west network reach. From that shell, they can query internal databases or call other backend services.
Native network rules and RBAC only check who connects or starts a session. They do not inspect what individual commands or queries do in flight. When a pod or engineer connects to an internal database, network rules see allowed packets. The cluster cannot tell a safe read query from a destructive write command. It also cannot mask sensitive fields like emails, phone numbers, or card numbers in results.
How to tighten internal traffic and cluster permissions
Start by mapping which services actually need to talk to each other. Apply default deny network rules to each namespace to block open pod communication. Next, add specific allow rules for known app dependencies and ports.
Review your RBAC bindings to remove broad cluster-wide permissions where possible. Use a Role and RoleBinding inside a single namespace instead of a ClusterRoleBinding. Separate routine log reading permissions from interactive pod execution permissions. This stops users from using pod shells as stepping stones across the internal network.
Check Kubernetes API requests and responses on the wire
Kubernetes RBAC decides who may call the API server. RBAC does not check response payloads or split listing Secret names from reading Secret values when a role grants get access. With hoop.dev, you run an open-source HTTP proxy, a sidecar for your Kubernetes API. The kubectl client changes only the server address it talks to while RBAC stays in place.
On each call, hoop.dev guardrails read the method, normalized path, and headers to refuse bad deletes or full Secret reads with a 403 error. On the way back, hoop.dev data masking walks JSON responses by key and masks matched values in Secrets, ConfigMaps, and pod logs. Every request and verdict is logged without storing bearer tokens. View the project on GitHub.
Frequently asked questions
What is the difference between north south and east west traffic in Kubernetes?
North south traffic moves between external users and the cluster boundary. East west traffic moves between pods and services inside the cluster itself.
Can pods in different namespaces talk to each other by default?
Yes, Kubernetes allows all pods to talk across all namespaces by default. You must add network rules to block or limit traffic between namespaces.
How does Kubernetes RBAC affect internal cluster security?
RBAC controls which users and service accounts can use the cluster API and subresources. Strict Roles and RoleBindings stop users from running shells inside pods to reach internal services.
Related guides
In this post