Skip to content
This repository has been archived by the owner on May 11, 2024. It is now read-only.
/ nodus Public archive

Simulated large clusters for Kubernetes scheduler validation.

License

Notifications You must be signed in to change notification settings

intel/nodus

Repository files navigation

DISCONTINUATION OF PROJECT

This project will no longer be maintained by Intel. Intel has ceased development and contributions including, but not limited to, maintenance, bug fixes, new releases, or updates, to this project. Intel no longer accepts patches to this project.

nodus podens

P → Q, PQ

Simulated large clusters for Kubernetes scheduler validation.

NOTE: This project is in the process of migrating to the kubernetes-sigs GitHub organization.

nodus podens

npsim masquerades as many Kubelets. Define classes of nodes and how many of each you want in a few lines of yaml. When a scheduler binds pods to npsim's fake Kubelets, npsim pretends to run them. The pods' runtime and terminal phase are driven by pod labels.

nptest interprets a scenario config file provided by the user. The scenario specifies the faked behavior of nodes and pods during the run, and includes assertions to validate the scheduler's behavior.

quick start

Build binaries

$ make

Start k8s control plane services

$ make k8s-up

Build a kubeconfig that points to the local cluster

$ make kconfig

Run a test scenario

$ nptest --scenario=examples/simple/scenario.yml \
  --pods=examples/simple/pods.yml \
  --nodes=examples/simple/nodes.yml
INFO[0000] run scenario                                  name="cpu resource test"
INFO[0000] run step [1 / 25]                             description="assert 0 pods"
INFO[0000] run step [2 / 25]                             description="create 1 large node"
INFO[0000] run step [3 / 25]                             description="assert 1 large node"
INFO[0000] run step [4 / 25]                             description="create 2 small nodes"
INFO[0000] run step [5 / 25]                             description="assert 2 small nodes"
INFO[0000] run step [6 / 25]                             description="create 1 4-cpu pod"
INFO[0000] run step [7 / 25]                             description="assert 1 4-cpu pod is Running within 4s"
INFO[0002] run step [8 / 25]                             description="delete 1 4-cpu pod"
INFO[0002] run step [9 / 25]                             description="create 3 1-cpu pods"
INFO[0002] run step [10 / 25]                            description="assert 3 1-cpu pods are Running within 4s"
INFO[0004] run step [11 / 25]                            description="change 1 1-cpu pod from Running to Succeeded"
INFO[0004] run step [12 / 25]                            description="assert 0 1-cpu pods are Pending"
INFO[0004] run step [13 / 25]                            description="assert 2 1-cpu pods are Running"
INFO[0004] run step [14 / 25]                            description="assert 1 1-cpu pod is Succeeded"
INFO[0004] run step [15 / 25]                            description="assert 0 1-cpu pods are Failed"
INFO[0004] run step [16 / 25]                            description="change 1 1-cpu pod from Running to Failed"
INFO[0004] run step [17 / 25]                            description="assert 0 1-cpu pods are Pending"
INFO[0004] run step [18 / 25]                            description="assert 1 1-cpu pod is Running"
INFO[0005] run step [19 / 25]                            description="assert 1 1-cpu pod is Succeeded"
INFO[0005] run step [20 / 25]                            description="assert 1 1-cpu pod is Failed"
INFO[0005] run step [21 / 25]                            description="change 1 1-cpu pod from Running to Pending"
INFO[0005] run step [22 / 25]                            description="assert 1 1-cpu pod is Pending"
INFO[0006] run step [23 / 25]                            description="assert 0 1-cpu pods are Running"
INFO[0006] run step [24 / 25]                            description="assert 1 1-cpu pod is Succeeded"
INFO[0006] run step [25 / 25]                            description="assert 1 1-cpu pod is Failed"

More info on scenarios here

View test results and session statistics

$ open my-test-result.html

Tear down k8s control plane

make k8s-down

About

Simulated large clusters for Kubernetes scheduler validation.

Resources

License

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages