Changelog

What's new in PodSteer.

Every release, newest first. Feature releases run down the main line; patch releases branch off the release they fix. Assets, checksums and the SBOM for each are on GitHub.

  1. v0.6.0 Latest Feature release

    A topology map with observed traffic, application maps and logs, and a pod table that scales to 10,000

    • A Topology page: a namespace or the whole cluster as one map, with findings on the boxes
    • Select several namespaces at once, in every list and on the Topology
    • Observed traffic from Istio, Linkerd, Beyla, Caretta or Hubble metrics, with nothing installed
    • A pod table that stays quick at 10,000 pods, and Linux and Windows installers
    Full notes and upgrade guidance

  9. Where it started ·

    The idea: a Kubernetes client that tells you what is wrong

    Every Kubernetes client we used did the same thing: it listed the pods and left the conclusion to us. The CrashLoopBackOff was right there on screen; the reason was three commands away, in the logs of a container that had already died.

    PodSteer started from one question — what would it take for the tool to draw the conclusion? — and three rules that have held ever since: native rather than Electron, nothing to sign in to, and nothing on screen it cannot back with the cluster's own state.