Hands-on Lab·Certified Kubernetes Administrator
Endpoints and EndpointSlices
A Service is a selector; the EndpointSlice is the answer to it. Scale a Deployment and watch the list change, then add a Pod that never becomes Ready and see it appear in the slice marked not-ready rather than being left out.
Services and Networking Guide 48 of 103 Intermediate
- Kubernetes1.36.4
- Cluster4 nodes
- Runtimecontainerd 2.2.6
- CNICalico v3.32.1
- TimeAbout 30 min
Slice names carry a random suffix. The conditions on each endpoint are the part that matters.
| Server Name | IP Address | OS | Roles | CPU | RAM | HDD |
|---|---|---|---|---|---|---|
| CKA1001 | 192.168.0.175 | Ubuntu 26.04 LTS | Control Plane Node | 2 Core | 4 GB | 50 GB |
| CKA1001-NODE01 | 192.168.0.176 | Ubuntu 26.04 LTS | Worker Node | 2 Core | 4 GB | 50 GB |
| CKA1001-NODE02 | 192.168.0.177 | Ubuntu 26.04 LTS | Worker Node | 2 Core | 4 GB | 50 GB |
| CKA1001-NODE03 | 192.168.0.178 | Ubuntu 26.04 LTS | Worker Node | 2 Core | 4 GB | 50 GB |
This guide includes
Use this when a Service exists and traffic goes nowhere. This matters because a Service is only a selector - the EndpointSlice is the answer to it, and reading that answer tells you at once whether the problem is selection or readiness.
- finding the EndpointSlice created for every Service, named after it with a random suffix
- scaling a Deployment and watching the slice follow, recording which node each endpoint sits on
- seeing a not-Ready Pod listed with
ready=falserather than quietly omitted - creating a Service whose selector matches nothing - the most common broken Service there is
Before you start
- The Services guide, since everything here is about what sits behind one.
- The probes guide, for what makes a Pod Ready. That condition is what decides membership below.
- A cluster with at least two nodes.
-
One EndpointSlice per Service, created for you
-
Scale the Deployment, watch the slice follow
-
A not-Ready Pod is listed, not omitted
-
A Service whose selector matches nothing