Projects and overview

Three pages answer three questions:

All figures come from your organization's health checks and incidents.

How the figures are calculated#

Health of a project or endpoint:

Services → Overview#

Use it as the live status board. It refreshes every 30 seconds.

• Banner: either All systems operational, or how many endpoints are down or degraded and which ones, plus the number of checks in the last hour.

• Figures:

  • endpoints up;
  • availability over 24 hours;
  • response p95 over 24 hours;
  • checks in the last hour and how many failed;
  • open incidents.

  Click Endpoints up to show only the endpoints that are down. Click Open incidents to open the incident list.

• Status board: one row per endpoint, grouped by project (untick Group by project for one list). Problems come first. Each row shows:

  Column	Shows
  Endpoint	Status dot, name, type, when it was last checked. When down: for how long. An incident tag opens the open incident
  24 hours	48 bars of 30 minutes. Green: all checks passed. Amber: some failed. Red: most failed. Grey: not checked. Hover a bar for its time and checks
  Uptime	Over the 24 hours
  p95	Over the 24 hours, coloured green, amber or red
  Last	The last check's response time

  Click a row to open the endpoint. Filter by Down, Degraded, Up or Paused, or search by name, URL or project.

• Recent changes: what went down and what recovered (and after how long) in the last 7 days. Click one to open its incident.

• Slowest, 24 h: the five slowest endpoints. Click one to open its response times.

• Shortcuts to Response times, Apdex & anomalies, Network and Certificates.

Projects#

• Period: 7, 30 or 90 days. It is kept in the address bar (?days=90).
• Figures:
  • Projects, and how many are below target. Click to show only those.
  • Endpoints up, with how many are down or degraded. Click to show the projects with a down endpoint.
  • Uptime, compared with the previous period.
  • Incidents: in the period, open now, and in the period before.
  • Time to resolve, and time to acknowledge.
• Needs attention: a line naming every project that is down or below its target. Click a name to open that project.
• Filters: search; All, Down, Degraded, Healthy or Below target; and a sort order (needs attention first, lowest uptime, most incidents, priority, name).

Each project card shows:

• the project colour, name, priority and health;
• uptime against the target, with the trend;
• the error budget bar;
• one bar per day, green when the day met the target. Hover a bar for that day's uptime and incidents;
• endpoints up out of the total (with how many are down), incidents (with how many are open), MTTR and p95;
• the weakest endpoint, when one wasn't up the whole time.

Click a card to open the project: edit it, see its endpoints, add one. On hover, a card also shows Analytics (that project in Project Analytics) and Delete (administrators only).

Endpoint adds an endpoint and New project creates a project. Endpoints not in a project are listed under Orphan Endpoints, where you can move them.

Link to one project with /dashboard/projects?open=<project id>.

Project Analytics#

• Period (7/30/90 days) and CSV, which downloads the comparison table.

• Figures:

  • uptime, with its trend;
  • projects within target out of those measured. Click to sort by error budget;
  • incidents compared with the previous period;
  • open incidents, with how many are critical;
  • time to acknowledge;
  • time to resolve.

• Uptime per day: all projects together (dashed) and one line per project (up to 8). Click a name in the legend to hide or show its line.

• Incidents per day, stacked by severity.

• Projects compared:

  • uptime and its trend;
  • one bar per day;
  • error budget;
  • incidents and how many are open;
  • time to acknowledge and to resolve;
  • p95;
  • endpoints up out of the total.

  Click a column header to sort. Click a row to open the project panel.

• Incidents by severity, What opened them (health checks, logs or metrics) and Weakest endpoints across all projects.

The project panel shows:

• the project's health and priority, with a link to open it in Projects;
• uptime, incidents, time to acknowledge and time to resolve;
• the error budget, and Change target for members and administrators: 99%, 99.5%, 99.9%, 99.95% or any value from 50 to 99.999;
• uptime per day, with the target as a dashed line;
• its endpoints, weakest first: uptime and trend, p95, incidents, and one bar per day. Click one to open the endpoint;
• its incidents, with duration or open. Click one to open the incident.

Link to a project's panel with /dashboard/project-analytics?project=<project id>&days=30.

API#

days is 1–90 (default 30).

Limits#

• Uptime counts checks, not minutes. An endpoint checked every 30 seconds and one checked every 5 minutes both count as one endpoint in the project average.
• Health checks are kept for HEALTH_RETENTION_DAYS (90 by default). A 90-day period compares with the 90 days before only if that history still exists.
• Recent changes lists outages detected by health checks. Incidents opened from logs or metrics appear on the Incidents page.