Status API

The Status API returns what the status page shows: the overall state of the platform, every monitored service with its uptime over the last 90 days, the active incident if there is one, and the most recent past incidents. It needs no key, and its responses are never cached.

When the state behind the page cannot be read, the call answers 503 with a Retry-After header rather than a quiet UP.

Get the status#

Only /v1/status exactly, and only GET: a sub-path such as /v1/status/services answers 404, any other method 405, and an OPTIONS preflight 204.

curl https://api.cachegenie.com/v1/status

{
  "CODE": 200,
  "MESSAGE": "API Request Success",
  "DATA": {
    "CURRENT_STATUS": {
      "STATUS": "UP",
      "MESSAGE": "All systems operational"
    },
    "SERVICES": [
      {
        "NAME": "CDN",
        "STATUS": "OPERATIONAL",
        "UPTIME_90D": 99.99,
        "LOCATIONS": [
          {
            "NAME": "Alpha City, XX",
            "STATUS": "OPERATIONAL",
            "UPTIME_90D": 99.99,
            "LOCATION": "Alpha City, XX"
          },
          {
            "NAME": "Beta City, XX",
            "STATUS": "OPERATIONAL",
            "UPTIME_90D": 100.0,
            "LOCATION": "Beta City, XX"
          }
        ]
      },
      { "NAME": "Dashboard (NOC)", "STATUS": "OPERATIONAL", "UPTIME_90D": 100.0 },
      { "NAME": "API", "STATUS": "OPERATIONAL", "UPTIME_90D": 100.0 }
    ],
    "ACTIVE_INCIDENT": null,
    "PAST_INCIDENTS": []
  }
}

• CURRENT_STATUSObjectthe overall state: STATUS, one of the values below, and MESSAGE, the headline the status page shows, such as "All systems operational", or the title of the active incident.
• SERVICESArrayone entry per service on the status page, in display order, each with NAME, STATUS and UPTIME_90D: the percentage of checks over the last 90 days that did not find the service down, to two decimal places, or null for a service with no checks yet. A service made up of locations, which the CDN is, carries LOCATIONS: one entry per location with the same fields, plus LOCATION, the same place name again. Its own UPTIME_90D is the average of its locations.
• ACTIVE_INCIDENTObject or nullthe incident or maintenance notice in progress, in the same shape as a past incident, or null.
• PAST_INCIDENTSArraythe twenty most recent resolved incidents, newest first, each with TITLE, DESCRIPTION (plain text), TYPE (one of "System Maintenance", "Degraded Performance", "Partial System Outage" or "Major System Outage"), TIMESTAMP (Unix seconds, when it was published) and SERVICES (the names of the services affected).

The overall status#

CURRENT_STATUS.STATUS summarises every service:

• UP: nothing is down, degraded or in maintenance.
• DEGRADED: at least one service or location is down, but not all of them.
• DOWN: every service that has been checked is down. A service still waiting for its first check is not counted either way.
• MAINTENANCE: at least one service is in planned maintenance and nothing is down.
• INCIDENT: an incident or maintenance notice is published. Read ACTIVE_INCIDENT for the details and the per-service statuses for the live picture.

A service's status#

• OPERATIONAL: the service is up. A check that fails on its own leaves it up.
• DOWN: three checks in a row have failed, or we marked the service down. An outage counts against uptime from its first failed check; a failed check the ones after it do not confirm does not count. A location that is down is taken out of DNS automatically, so viewers are routed to the next nearest location.
• DEGRADED: some but not all of what is behind the service is down: some of a group's locations, or some of a location's capacity.
• MAINTENANCE: planned work, or a place we have taken out of service for now. Checks during maintenance do not count towards uptime.
• PENDING: the service was added recently and its state is not settled yet.
• PLANNED: a location we intend to serve from and have not opened yet. It reports no UPTIME_90D and never counts towards the status of the group it sits under.