Get started

Your first answer, in four steps.

Connect what you already have, review the first situations, and follow the use case that matches. Nothing changes on your network until you approve it.

  • Timeabout an hour to first situations
  • You needa NetPulse sign-in and read access to your telemetry

Do this

Four steps

  1. Connect your sources

    Start with device telemetry and link performance. Add inventory, tunnels, application quality and security events as you go; each one sharpens the picture.

    You should see
    Your sites, links and providers in the Explorer, each with live metrics.
  2. Let it learn what normal looks like

    The agents build a baseline for every link and device by hour of day and day of week, so a Monday-morning peak is not mistaken for a fault.

    You should see
    Detections that compare each metric with its own normal, not a fixed threshold.
  3. Review your first situations

    Related alerts arrive grouped as one situation with a named root cause, the evidence behind it and the business impact. Open the timeline and read the story.

    You should see
    One situation per problem, each with a root cause and a timeline.
  4. Follow the use case that matches

    Each use case is a short, ordered path through the same screens, from the first signal to a verified fix.

Before you connect anything

What can the agents see?

Eight things the agents use. Tick what you have to see which gaps would limit them. It runs in your browser and sends nothing anywhere.

0%Tick what you have
    A checklist of inputs, not an assessment of your environment.

    Before you act

    Check the call first.

    The root cause NetPulse names is a hypothesis with evidence attached. Four checks take under a minute and tell you how hard to lean on it.

    1. 1

      Read the evidence

      Open the situation’s timeline. Do the onset, the affected sites and the suspected root tell one consistent story? If you cannot explain it in a sentence, do not act on it.

      Situations
    2. 2

      Look at the track record

      Accuracy shows how often the root-cause call, and the forecast’s lead time, have been right against known incidents, by difficulty and by entity type. Where a class is less certain, have a person confirm.

      Accuracy
    3. 3

      See what changed

      Open the change record around the onset. A firmware upgrade or config edit shortly before the first signal is often the cause, and the situation lists it for you.

      Situations
    4. 4

      Approve with the blast radius in view

      Every proposed fix states what it touches and how recovery will be verified. Approve it, or hold it and take the recommendation into your own change process.

      Approve or hold a fix