Glossary
Terms are listed alphabetically. Where a term has a name in the API, it is shown in code.
Alert delay - How long a failure must last before a particular contact is notified (alertDelay, in minutes).
Used to build escalation ladders.
Alert subscription - The link between a monitor and a contact that says which state changes (up, down,
repeatedlyDown) the contact is notified about. See Subscriptions.
API token - A bearer credential you mint under Integrations -> API; it carries scopes, an expiry and an optional IP allow-list. See API authentication.
Assertion - A rule that checks part of a response (status isOk, body contains "Checkout"). All of a monitor’s
assertions must pass. See the assertion language.
Attached sub-check - An SSL expiry, domain expiry, DNSBL or Web Risk check that runs alongside an HTTP monitor
instead of as a separate monitor (attached).
Band - The status page tier your plan falls into, from the number of monitors it includes. See status page plan limits.
Check - One run of a monitor’s test from one location at one moment.
Check log - A monitor’s recorded results (GET /monitor/{id}/result). Consecutive checks with the same outcome
are grouped into one row.
Codename - The short, stable name for why a check failed (Timeout, DnsHostNotFound, Http 503), in
error.codename. See error codes.
Component - One row on a status page, backed by a monitor or set by hand (third-party).
Confirmation - The code a new email, SMS or voice contact must enter before it receives alerts.
Contact - A place alerts or reports go: an email address, phone number, messenger chat, webhook URL and so on
(contact).
Contact group - A named set of contacts, each with the events it receives, used to subscribe several contacts to monitors at once.
Cron schedule - A cron expression that runs a monitor at set times instead of every N seconds
(cronSchedule).
Down detection / recheck - How HostTracker confirms a failure from other locations before recording a monitor as
down (recheck.strategy). See down detection.
Error budget - The downtime an SLA target allows over a period; 99.9% over 30 days is about 43 minutes.
Fallback - What happens when no location in a monitor’s pools is available: geo, world or starve.
Full log - A monitor option that logs checks in 5-minute groups instead of 60-minute groups (fullLog).
Idempotency key - A header (Idempotency-Key) that makes a retried API write return the first result instead of
running twice.
Incident - A confirmed down episode of a monitor, from the confirmed failure to the confirmed recovery. See What is an incident. Not the same as a status page incident.
Instant check - A one-off check you run now, outside any monitor’s schedule (POST /check). See
Instant checks vs monitors.
Interval - How often a monitor runs, in seconds in the API (interval).
Job - An asynchronous API operation (bulk create, report generation) you poll with GET /job/{id}.
Location (agent, checkpoint) - One monitoring vantage point with a country and city (GET /agent).
Maintenance window - A scheduled period in which chosen monitors hold back alerts and/or statistics. See maintenance.
MCP - The Model Context Protocol; HostTracker’s MCP server lets AI assistants operate your account. See MCP.
Monitor - A configured, recurring check of one target, with a type, schedule, locations and subscriptions
(monitor).
Monitor state - up, down, paused or maintenance.
Package (plan) - Your subscription tier; it sets your limits. See plan limits.
Pool - A named group of locations (a region, or allworld for everywhere) used to choose where a monitor runs.
Public statistics - An option that makes a monitor’s statistics viewable without signing in (openStat).
Report subscription - The link between a monitor and an email contact that receives scheduled uptime reports
(daily to yearly).
Scope - What an API token may do, per area (monitor:read, contact:write, …).
Severity - An incident’s size by duration: minor (under 5 min), major (5-60 min), critical (1 hour or
more).
SLA target - The uptime percentage you commit to (slaTarget). See Uptime and SLA.
Slug - The permanent part of a status page’s address: status.host-tracker.com/<slug>.
Snapshot - What a failing check received: an HTTP response or a browser screenshot. See snapshots.
Status page - A public (or password-protected) page showing your services’ status to your users. See status pages.
Status page incident - An announcement you declare on a status page, with a timeline of updates, sent to its subscribers.
Subaccount - A separate login with limited rights on your account. See subaccounts.
Subscriber - Someone (email) or something (Slack, Teams, webhook) that receives a status page’s announcements.
Tag - A free-form label on a monitor, used to filter and to target bulk edits and webhooks.
Third-party component - A status page row for a service you do not monitor, whose state you set by hand.
Uptime - The share of measured time a monitor was up over a period, excluding downtime inside maintenance windows that suppress statistics.
Webhook - A signed HTTPS delivery of HostTracker events to your endpoint (webhook). See
webhooks.
Webhook contact - An alert contact of type http that posts monitor alerts to a URL, with an optional custom
payload. Not signed; different from a v2 webhook.