Documentation
Documentation
What IgniteOps is made of, how the pieces fit, and a page for each of them with every option, every limit and the matching API call.
How the pieces fit
An alert travels through five things, each of which you configure once and the API exposes in full:
An alert arrives as one HTTP POST, from a monitoring tool, a script, a flow tool or a person. Repeats with the same alias update one alert.
A routing rule matches it on severity, source, title, tags or custom fields and picks who handles it, usually an escalation policy.
An escalation policy pages its targets step by step, minutes apart, until someone acknowledges, and can repeat.
A schedule says who is on call at that instant, from rotations and overrides, in the team’s time zone.
Notification methods reach that person the way they chose: push, email, SMS and a phone call, in their order and with their delays.
Teams own the rules, policies, schedules and keys. The organization owns the members, the seats and the plan. Everything the console does is also in the REST API, documented from the same OpenAPI document the API is deployed from.
Where to start
- Evaluating it: read escalation policies and schedules, which are where products in this category differ, then plans and limits for every number.
- Setting it up: getting started takes you from an empty organization to a test alert on your phone, and integrations has the configuration for Datadog, Grafana, CloudWatch, GitHub Actions and Alertmanager.
- Not an engineering team: the no-code guide sends alerts from Zapier, Make, Power Automate or n8n and drives a schedule from an Outlook calendar.
- Security review: the security page describes how the service and the credentials are protected, and the roles page has the permission matrix.
Each page says when it was last checked against the product. If something on a page disagrees with what the console does, the console is right and we would like to know: [email protected].
Every page
Start here
Configure
- Alerts and ingestThe payload, deduplication by alias, the alert’s life, quotas and the metrics computed from alerts.
- Routing rulesConditions, operators, priority, targets, and what happens when nothing matches.
- Escalation policiesSteps, delays, the eight target types, repeats, and what stops an escalation.
- Schedules and overridesRotations, time zones, active hours, weighted turns, overrides, who is on call and the .ics export.
- Notification methodsHow each person is reached: push, email, SMS and voice, their order and their delays.
- Teams and rolesOrganizations, teams, the two sets of roles and what each one may do.
Connect
- DeploymentsRecording deployments from a pipeline, their statuses, the team list and timeline, and how the insights read them.
- IntegrationsRecipes for Datadog, Grafana, CloudWatch, GitHub Actions, cron and Alertmanager, and the outbound webhook.
- The REST APIThe three credentials, the base URL, idempotency, and the OpenAPI document.