CASE STUDY · LIBRENMS

The features were already there. Copilot found them.

An operator had capable monitoring in place, but useful features were buried under the noise of alerts and dashboards. Network Copilot reviewed their setup and surfaced five reporting capabilities they weren't using yet.

Operator
Network operator, East Africa
Monitoring
LibreNMS
Equipment
Huawei and MikroTik
Focus
Reporting and capacity
A network operations engineer reviewing Network Copilot's recommendations on a laptop.
01

The challenge

The operator's monitoring was doing its job. LibreNMS was collecting data from devices across the network, and alerts were reaching the team. But day-to-day work was dominated by dashboards and alert queues, so the features that turn that data into reports were left unexplored.

The tools were capable. They were already installed, and the data was already being collected. What was missing was the time to find out what else they could do.

02

What they asked

Instead of working through documentation and menus, the team asked Network Copilot to review their monitoring setup and point out what they weren't getting from it.

Review our LibreNMS setup. Which useful features are we not using yet?

I've reviewed your configuration. Here are 5 reporting features you aren't using yet, and why each one matters for your network.

Paraphrased from the operator's conversation with Network Copilot.
03

What Copilot found

Copilot's review covered several areas of the setup. This part focused on reporting: five features already available to the operator that had not been switched on.

  1. 01 Not in use

    Port groups and description tags

    What it is
    Label ports with a consistent description convention, such as Cust: Acme, Transit: Upstream-A or Core: site-01, and group them.
    Why it matters
    Report bandwidth per customer and per upstream provider, and limit alerts to the ports that matter.
  2. 02 No bills set up

    95th-percentile and quota billing

    What it is
    Set up bills on transit, internet exchange (IX) and major-customer ports, using 95th-percentile or quota-based calculation.
    Why it matters
    Produce monthly usage and capacity reports from data LibreNMS is already collecting.
  3. 03 Depends on earlier fixes

    Availability by device group

    What it is
    Group devices by customer or service and report availability for each group.
    Why it matters
    Turn existing uptime data into customer-facing SLA reports.
  4. 04 Not in use

    Scheduled maintenance windows

    What it is
    Schedule maintenance in advance for the devices and services affected by planned work.
    Why it matters
    Planned work no longer counts against availability or triggers unnecessary alerts.
  5. 05 Check if in place

    Configuration backups with Oxidized

    What it is
    Connect Oxidized to LibreNMS, if it is not already in place, to back up configurations from the Huawei and MikroTik devices.
    Why it matters
    Keep a history of configuration changes and a backup to restore from.

Availability reports only become accurate once the earlier fixes in Copilot's review are complete, so Copilot flagged that dependency instead of recommending the feature on its own.

Features are hidden in the noise of alerts and dashboards. Network Copilot brings them out, helping you make maximum use of your existing infrastructure.

04

Why it matters

  • More from what you already have. Four of the five recommendations are features already built into LibreNMS. The fifth connects an existing open-source tool.
  • Reports customers and managers can use. Per-customer bandwidth, monthly billing data and SLA availability all come from data that was already being collected.
  • Less noise for the team. Port groups and maintenance windows mean fewer unnecessary alerts, so engineers can focus on the ones that matter.

Operator details have been anonymised, and customer and provider names in examples have been replaced. Network Copilot's findings depend on the data, integrations and permissions available in each environment, and recommendations should be validated by the responsible engineer.