Book a demo

NOC moduleNew

The network warns you,
not the other way round

The NOC watches, on its own, the routers, nodes, links, access points and OLTs you already have on the system: nobody has to tell it what to look at. And it does not stop at warning you about what has already dropped — it shows you the fibre going bad, the box running out of ports and the uplink heading for saturation, with the estimated date.

See it on your network See the whole platform

The NOC's Live panel: counters for down and flapping equipment, ONT signal, network latency, outages over the last 24 hours and the list of faults with their affected customers — The first thing you see. The counters on top; underneath, each outage with the equipment it drags down, the technician assigned and the ticket opened. It is not just how many devices are down: it is also how manycustomersare behind them and how much they bill each month, which is what decides where you start.

Before the fault

See what is about to break

Anyone can see an outage: the customer calls. What costs money is what is slowly going wrong and nobody is looking at.

Fibre signal anomalies

ONTs with an unstable signal or intermittent micro-drops: the loose connector or the badly made splice that in two weeks will be a fault and a site visit.

Optical health of the plant

It compares each connection's real signal against its theoretical optical budget and flags the ones losing too much. Plus power by area, to see when a whole PON port drops.

Predictions with a date

OLT temperature and CPU, uplink usage: the NOC projects the trend and tells you what is heading for its limit andwhen it will get there.

Anomalies with AI

Each metric learns how it normally behaves at each hour of the day. A node carrying four in the morning's traffic at four in the afternoon breaks no threshold, but something is wrong with it.

Blind spots

And what you cannot see by looking at the network

Internet health

If your public IPs land on a blacklist, your customers' e-mail starts bouncing and nothing looks wrong on your network. The NOC watches them every night, and that your DNS keeps answering too.

Botnet traffic

Subscribers with the traffic patterns of an infected device: sending spam directly, connections to abuse ports or port sweeps. That is what ends up putting your IPs on those blacklists.

Plant capacity

Usage of boxes and PON ports, the fullest first, and a warning when a box is nearly out of room. So you do not sell coverage where nobody else fits.

Network log

The outages and recoveries the equipment writes to syslog, without the noise of everything else.

When you have to answer

Paperwork you have no idea
where to get

A business customer demands the month's SLA. A court order arrives asking who was using an IP at a specific time. Two pieces of paper that cost an afternoon each, and the NOC issues them on its own.

Availability and credits

How long each customer has been out of service because of faults the NOC itself detected, and the proportional part of their fee that would correspond. Overlapping outages — the node drops and its access point with it — count only once.

The credit is a reference: no credit note is created and no invoice is touched. And the report can be scheduled to go out by e-mail every day, week or month.

Availability screen: customers with any outage, time out of service, availability percentage, monthly fee and proportional credit

CGNAT traceability

With CGNAT many customers go out through the same public IP, so the question «who was using IP X, port P, at such a time?» cannot be answered by looking at the router. Here you type the IP, the port and the time — local or UTC, as the requests usually arrive — and the subscriber comes out.

Underneath the result you put the file number and the authority, and the PDF report is generated. If there is no result, the report says so: that is a valid answer too.

Legal PDF report identifying a subscriber by public IP address and port, with the request, the data consulted in local time and UTC, the result and the method

Legal retention

Public IP to subscriber translations are kept compressed for as long as you configure — 365 days by default — consolidated by session, not by minute.

With a fingerprint

Each report carries its sequential number, a fingerprint of its data and a copy of the PDF, so you can prove afterwards that it has not been altered.

Maintenance windows

You plan the works window and during those hours the chosen equipment does not raise alerts. Planned works stop looking like a fault.

Out of the office

The state of the network,
without opening the computer

Las averías no esperan a que estés sentado. El NOC se consulta desde el móvil, y si prefieres no entrar siquiera, se le pregunta al bot de Telegram: escribes /estado o /caidos y contesta al momento, solo a quien esté vinculado.

Everyone builds their own panelThey move the widgets, resize them, hide the ones they never look at and add the ones they are missing. It is saved for them, not for the company.
Equipment that is not yoursSNMP v1, v2c and v3 monitoring of equipment outside the inventory: interface drops, traffic, CPU, memory, temperature, fans and power supplies.
And on the mapThe NOC's status is drawn on GeoRed: what is down and what it drags with it, over the real plant.

The NOC's status on the GeoRed map: equipment down in red, flapping in amber and in service in green, over the fibre plant

How it does it

Nobody has to tell it what to look at

The NOC takes the routers, nodes, switches, links, access points and OLTs you already have on WispControl and checks them by itself. Each at its own pace:

What	How it is read	How often
Network equipment	Ping	1 minute
MikroTik routers	Router API	2 minutes
OLTs, PON ports, boxes and ONTs	SSH to the OLT (Huawei, ZTE) or the FELIX API (Alea)	5 minutes
SNMP equipment and access points	SNMP v1, v2c or v3	5 minutes
Internet health and optical health	Checks from the server	Every night

And it does not multiply the messages: when a node drops and everything hanging off it with it, the alert talks about the cause and the rest appear as dragged down.

How many of these can you see on your network today?Half an hour, with your equipment and your OLTs in front of you.