# Uptime Monitoring — psiLynx

> Track page availability around the clock. psiLynx checks your URLs at regular intervals, detects outages and content anomalies, and sends email alerts when something changes.

## Overview

Uptime Monitoring runs independently from PageSpeed checks. Monitors are configured per URL within each project. One project supports up to **5 monitors**.

| Status | Meaning |
|---|---|
| **UP** | URL responds with HTTP status < 400 |
| **DOWN** | URL is unreachable or returns an error |
| **DEGRADED** | Server returns 200 OK but page size dropped more than 50% below baseline — likely serving a stub or error page |
| **Unknown** | Monitor has not completed its first check yet |

DEGRADED catches cases where the server is technically responding but delivering incomplete content — useful for detecting maintenance pages or deployment issues that don't return an error code.

## Status badge

The current status is visible in two places:

- **Project header** — a UP / DOWN / DEGRADED badge next to the project name. Clicking it opens the Uptime tab. Updated on every tab switch.
- **Projects list** — a badge next to each project name.

The badge is hidden when no monitors are configured or all monitors are inactive.

## Uptime tab

Open a project and click the **Uptime** tab.

### Tab header

- Counters for UP (green) and DOWN (red) monitors
- Slot usage indicator: X / 5 monitors

### Monitor card

| Element | Description |
|---|---|
| **Status dot** | Colour indicator: green / red / yellow / grey |
| **URL** | Address of the monitored page |
| **Tag** | UP / DOWN / DEGRADED / Unknown |
| **Interval** | How often checks run |
| **Baseline** | Reference page size in bytes — recorded on first successful check |
| **Toggle** | Enable / disable the monitor without deleting it |

## Availability chart

Visual uptime history with a time-range switcher:

| Range | Granularity |
|---|---|
| **Recent** | Last 60 checks (one bar per check) |
| **Day** | By hour — 24 bars |
| **Week** | By day — 7 bars |
| **Month** | By day — 30 bars |
| **Year** | By ISO week — up to 52 bars |

Bar colours: green ≥ 95% uptime · yellow 80–94% · red < 80% · grey — no data.

## Events table

Detailed log of individual checks:

| Column | Content |
|---|---|
| **Time** | Date and time of the check |
| **Status** | UP / DOWN / DEGRADED |
| **HTTP code** | Server response code |
| **Response time** | Time to response in ms |
| **Size** | Response body size in bytes |
| **Error** | Error message if the check failed |

20 events are shown by default. Click **Show N more** to load additional rows.

## Adding a monitor

1. On the Uptime tab click **+ Add monitor**.
2. Select the URL to monitor from the list of active project URLs.
3. Set the **check interval** (in minutes).
4. Set the **timeout** — maximum wait time per check (in seconds, default 10).
5. Configure notifications:
   - **Notify on down** — email when status changes to DOWN or DEGRADED
   - **Notify on recovery** — email when status returns to UP
6. Click **Add**.

**Notes:**

- Each URL can only have one monitor per project.
- The project limit is 5 monitors. The Add button is disabled when the limit is reached.
- The page size baseline is set automatically on the first successful check.

## Managing monitors

### Enable / disable

Use the toggle on the monitor card. Disabled monitors stop running checks but retain their history and settings.

### Delete a monitor

Click the trash icon on the monitor card and confirm.

> Deleting a monitor permanently removes all its event history.

## Email notifications

Alerts are sent to your account email when a monitor's status changes:

| Event | Email |
|---|---|
| UP → DOWN | Site is Down |
| UP or DOWN → DEGRADED | Suspicious Page Change — server responds 200 but page content shrank significantly; manual review recommended |
| DOWN or DEGRADED → UP | Site Recovered |

Each notification type is controlled independently via **Notify on down** and **Notify on recovery** in the monitor settings.

## Data retention

- Raw check events are kept for **30 days**, then aggregated and removed.
- Daily aggregates are retained for up to **1 year** and power the Week / Month / Year chart ranges.