← Back to Kubernetes Resource Limits Check
Polaris, maintained by Fairwinds, is an open-source policy engine for Kubernetes covering 30+ built-in best-practice checks -- resource requests and limits among them, alongside security context, health checks, image tag pinning, networking, and more -- with the option to add custom checks defined in JSON Schema. It is deployed: as a dashboard that scores a cluster against its policies, as an admission controller that can reject or modify a non-conforming workload before it is ever created, or as a CLI wired into a CI/CD pipeline to check local YAML files.
This checker answers one narrower question with no deployment step at all. It is a zero-dependency,
zero-install browser page and a single Node file: paste or load workload YAML and get a report of
which containers are missing a memory limit, memory request, CPU request, or ephemeral-storage limit,
which have a request set above their own limit, and which set a CPU limit worth being aware of. It has
no admission-controller mode, no cluster dashboard, and nothing to deploy: node cli.mjs check
manifests.yaml.
| What you are trying to do | Where to do it |
|---|---|
| Enforce resource-limit policy (and 30+ other best practices) continuously across a live cluster, rejecting non-conforming workloads at admission. | Polaris -- that is its built-in admission-controller mode. |
| Score an entire cluster's existing workloads against a broad policy set and track the score over time. | Polaris -- that is its dashboard mode. |
| Check security context, image tags, health checks, or networking config alongside resource limits. | Polaris -- this checker covers resource limits only, deliberately. |
| Check one manifest file with nothing to deploy or install, and nothing uploaded anywhere. | This checker -- open the page, paste or load the file, done. |
| Add a CI gate that is a single committed file with no external binary or cluster connection to set up. | This checker -- cli.mjs is the whole dependency. |
| Understand why a CPU limit specifically can throttle a container even on an idle node, before deciding whether to set one. | This checker's own explanation page -- Polaris flags the field but does not walk through the CFS mechanism. |
These compose rather than compete. A team already running Polaris as an admission controller loses nothing by also keeping this narrower, install-free checker around for a contributor to run against a manifest before it is even proposed -- catching the same class of gap earlier and for free. And a team that does not want to deploy a dashboard or an admission controller just to check resource requests and limits on a handful of manifests can reach for this instead, with nothing to set up at all. Neither replaces the other; they trade breadth and continuous enforcement for zero deployment cost in opposite directions.
They sit at different scopes. Polaris is a policy engine covering 30+ Kubernetes best-practice checks -- resource requests/limits among them, alongside security context, image tags, health checks, and more -- deployed as a dashboard, an admission controller, or a CLI wired into CI/CD. This tool checks one narrow thing: whether a workload's memory limit, memory request, CPU request, and ephemeral-storage limit are present, plus whether a request exceeds its own limit. If you want broad, ongoing policy enforcement across a cluster or a whole manifest repository, Polaris is built for that. If you want a single YAML file checked right now with nothing to deploy or install, this is built for that instead.
For the resource-requests/limits question specifically, yes -- both are checking the same underlying Kubernetes fields. Polaris covers that as part of a much broader policy set (30+ built-in checks spanning security, networking, health checks, and image configuration, plus custom checks you can define with JSON Schema), and can enforce it continuously as an admission controller so a non-conforming workload never lands in the cluster at all. This tool covers only the resource-limits question, with no deployment step: paste or load a manifest and get a report immediately.
Because they cost different amounts of setup for a narrow, one-off question. Polaris is deployed infrastructure -- a dashboard or admission controller running in or against your cluster, or a CLI you install and wire into a pipeline -- which is exactly the right shape for continuous, cluster-wide policy enforcement. This tool is one dependency-free JavaScript file: node cli.mjs check manifests.yaml, nothing to deploy, nothing to install beyond Node itself. If a team already runs Polaris for its broader policy coverage, this does not replace that -- it is a lighter option for a single manifest someone wants checked before it ever reaches Polaris or a cluster at all.
No. This site is not affiliated with, endorsed by, or sponsored by Polaris or Fairwinds, named here only to describe what this checker is and is not. What Polaris currently checks and how it is licensed are described on its own documentation and repository -- check there for what it covers today.
Only point either tool at a manifest or cluster you own or are authorised to test. That rule is unchanged by which one you pick.