LineaChiara

When the Internet goes wrong,
where do you start?

One number cannot describe your entire connection. A line can download quickly but struggle with video calls; one website can fail while the rest of the Internet works. The goal is to collect evidence that distinguishes these situations.

The four measurements, without shortcuts

Download and upload describe how much data arrives or leaves per unit of time. Mbps means millions of bits per second; MB means millions of bytes, with eight bits in each byte. A browser measurement describes the path to one particular server. Your advertised plan may describe a theoretical maximum: check the guaranteed minimum and applicable conditions separately.

Latency is the time needed to receive a response. Here we measure HTTP requests, so the result includes browser and server work. It is not identical to an ICMP ping or the number shown in a game.

Jitter describes how response times vary. LineaChiara calculates the mean absolute difference between consecutive valid samples to the same destination. Timeouts break the sequence. Different methods produce different numbers: compare sessions using the same method.

Why measure during a transfer?

When uploads or downloads occupy a connection, other requests may have to wait. A much higher median latency under load is compatible with queueing, often called bufferbloat, but can also involve Wi-Fi, the device or the server. A small data budget may not saturate a fast connection. In that case, a small increase does not rule out a problem.

Change one thing at a time

  1. Record the symptom. Which application, at what time? All websites or just one? Record the situation before restarting equipment.
  2. Measure under normal conditions. Note Wi-Fi, VPN, mobile or Ethernet use. Record active backups, updates, streaming or other users on the network.
  3. Repeat over Ethernet if possible. Keep the device and server the same. If the device changes too, note that: the comparison isolates fewer variables.
  4. Compare another device and destination. A problem in one browser suggests different checks from one affecting the whole home.
  5. Repeat when the symptom occurs. Three tests at different times can be more informative than ten identical tests in a minute. Use the history and retain failures too.

If latency increases under load even over a cable, some routers offer SQM or QoS queue management. Check the model’s documentation, save the current configuration and compare one change at a time. Avoid factory resets, purchases or DNS changes based on a single test.

When the Internet is completely down

A public web app cannot load for the first time without a connection. After a complete first visit, LineaChiara tries to cache its files for offline use. Browser support, available space and cache retention matter. Load it before an outage, or keep a locally runnable copy.

During an outage you can record symptoms, run monitoring and export a report of failed requests. Speed remains unavailable when measurement servers cannot be reached; the app does not invent a zero. “Browser online” does not prove Internet access: some systems only check for a local connection.

Record the modem or optical network terminal indicators and timestamps. Do not unplug or handle the optical connector. A loss-of-signal indicator, a persistent total outage across devices, or recurring Ethernet disconnections are reasonable reasons to open a support ticket. Interpret indicator lights using the equipment manual.

A report that support can act on

Provide a precise time window, connection type, the apps or websites you could not use, tests actually performed and samples. Ask for a ticket number, the checks performed and the next update. “It does not work” is hard to correlate; “between 20:14 and 20:19, two wired devices could not reach multiple destinations” is an investigable observation.

Restarting can be a useful test, but removes context and does not prove a lasting fix. Record its time and compare before and after. If asked to wait, ask what event or work is expected and how to verify recovery. Precise details are more useful than aggressive language.

These reports are diagnostic material, not a certification. For contractual disputes in Italy, consult MisuraInternet and Ne.Me.Sys. and the current complaint procedure. Elsewhere, check the relevant national regulator and your provider’s process. Do not assume the Italian procedure applies internationally.

Three conclusions to avoid

  • “The router does not answer ping, so it is broken”: ICMP may be filtered or deprioritized.
  • “The test is slow, so the provider is responsible”: the device, Wi-Fi, VPN and server are part of the measurement.
  • “An intermediate hop loses packets”: if later hops respond normally, an intermediate router answering traceroute infrequently does not prove loss of forwarded traffic.

A browser cannot freely read local routes, Wi-Fi radio quality, DSL signal-to-noise ratio or attenuation, optical power, modem counters, path MTU, listening ports or router logs. The Windows extra covers some local checks; modem information requires the equipment’s diagnostics or provider support.

Collect your first measurements

App method v0.3 · Documentation updated 17 September 2026.

Single, timed or continuous tests

A single test measures download and upload for 5, 10, 20 or 30 seconds each, in addition to initial checks. In-flight requests may take up to 6 extra seconds to finish. Timed observation lasts minutes, hours or days, with 24-hour, 3-day and 7-day presets plus a custom duration; “No end time” runs until Stop. Keep the page open and the device awake: browsers can suspend background work.

Full network repeats speed measurements at your chosen interval and observes stability between them. Stability only makes small HTTP requests every 5, 15, 30 or 60 seconds, according to your choice, without measuring download or upload speed. Speed tests never overlap. Time-based modes have no automatic MB cap and may use many GB. An optional data budget limits each speed measurement but does not end the observation.

Export interim reports without stopping observation. Memory retains at most 10,000 samples, 100 speed cycles and 2,000 recent events. Reports disclose discarded evidence: statistics describe the retained window, while traffic totals cover the entire session. History, cache and reports do not constitute an always-on monitoring service.

Hours, days and missing responses

No end time continues until you press Stop. Timed mode supports minutes, hours, days or a custom duration. Stability only is recommended for a full day: small requests without continuous large transfers. Full network can repeat speed measurements every 15, 30 or 60 minutes.

Hourly history records successful and failed requests, mean resting latency, jitter and speed cycles completed in that hour. Up to 2,160 hourly buckets and 1,000 periods are retained independently of 10,000 raw samples, 100 speed cycles and 2,000 recent logs. Reports disclose removed evidence. Missing hours are not filled with zeros and intermittent checks do not establish continuous uptime percentages.

We compare the selected server with Google and distinguish missing responses on one or both destinations. This cannot identify provider responsibility. Timers delayed or suspended for over 30 seconds produce a period without observations; recovery is confirmed only by a later successful check. Closing the page or powering off stops measurements. The screen-awake option depends on browser support and does not prevent system sleep.

If local retention is enabled, a checkpoint is saved at startup and approximately every minute of activity, when supported by the browser. Reopening can recover data up to the latest save. It does not invent missing measurements or restart traffic automatically. A normal stop saves to history. Download reports and JSON to archive long observations.

PDF reports ready to share

User and technical reports download directly as PDFs, branded with links to PHABDEV. HTML and Markdown are alternatives. Technical PDFs embed session.json and samples.csv. Compatible readers expose them in the attachments panel; if a browser hides attachments, download JSON and CSV separately. Your IP selection also applies to embedded files; free-text notes are not automatically redacted.

Alternative servers

Cloudflare remains the default. Compare it with the public Clouvider servers in Frankfurt and London listed in the official LibreSpeed directory. Shared resources: availability, capacity and conditions can change. No PHABDEV partnership is implied. We send no reports, notes or telemetry, but operators see diagnostic requests and your IP. The adapter uses garbage.php downloads in 1 MiB chunks and empty.php for uploads and latency. Different destinations can produce different results on the same connection.

Guides by problem