How a Device Problem Becomes a Fix

What this covers

Everything so far has started with a person asking for something. This one starts with a machine noticing its own problem — ideally before the user does.

This matters most if you look after IT equipment, either your own or your clients'.

The flow

Agent installed  →  Reports in  →  Threshold crossed  →  Alert  →  Ticket  →  Fixed

1. The agent goes on the machine

A small program — the agent — is installed on each computer or server. It runs quietly and reports back. It works on Windows, macOS and Linux.

Without an agent, Plenix cannot see the machine. This is the step people skip and then wonder why a device shows nothing.

2. It reports what it sees

Continuously, and without anyone asking: disk space, memory, whether it is switched on, what software is installed, which updates are missing, and whether backups and antivirus are working.

3. Something crosses a line

You set the lines: "tell me when a disk is over 90% full." When one is crossed, an alert is raised.

Alerts have severity, so a full disk on a server outranks one on a spare laptop.

4. The alert becomes a ticket

This is the important join. An alert nobody turns into work is an alert nobody actions. Plenix can raise a ticket from an alert automatically, so it enters the same queue as everything else and gets the same SLA clock.

5. Somebody fixes it

Often without leaving Plenix:

• Remote control — take over the screen, with the user watching.
• Run a script — clear temporary files across every affected machine at once.
• Push updates — approve and deploy missing patches.
• Send someone — if it needs hands, raise a work order.

6. It closes itself, sometimes

If the fix worked, the next check-in shows the disk healthy and the alert clears. With automated remediation, common problems can be fixed and closed before anyone has read the alert.

Why this is worth setting up

The difference between reactive and proactive support is this flow. Without it you find out a server is full when it stops working and someone phones you, angry. With it, you cleared it last Tuesday and nobody noticed.

For your IT team

• Agent is a Go binary, cross-platform; remote control uses WebRTC on Windows, X11/Wayland on Linux, and macOS where enabled.
• Patch management, software inventory and script execution are per-device or per-group.
• Automation policies can remediate without human approval — scope them carefully before enabling.

Where to go next

• What happens once it is a ticket → How a Request Becomes a Resolved, Billed Ticket
• Printers specifically → Printer Monitoring
• Hands-on-site work → Field Service