Trigger conditions
The components that start crawl runs on a schedule or in reaction to events.
Trigger conditions create new runs automatically, either proactively on a schedule or reactively when something happens. They are configured as job triggers, not in the manifest config. This page catalogs the available kinds and their parameters.
How triggers are configured
A trigger is attached to a job and has a name and a condition. The condition is an object with a kind and, for some kinds, a params object. Triggers are created through the orchestrator API; see Schedule recurring crawls for the workflow. A schedule condition is proactive (it fires on a clock); an on_finish condition is reactive (it fires when a run finishes).
{
"name": "nightly-refresh",
"condition": {
"kind": "schedule",
"params": { "expression": "0 30 2 * * * *" }
}
}schedule
Fires on a cron schedule.
| Field | Type | Required | Default | Description |
|---|---|---|---|---|
expression | string | yes | none | A seven-field cron expression, evaluated in UTC. |
The seven fields are second, minute, hour, day-of-month, month, day-of-week, and year. For example, 0 30 2 * * * * fires at 02:30:00 UTC every day. Because the schedule is evaluated in UTC, convert your intended local time to UTC when writing the expression.
{
"kind": "schedule",
"params": { "expression": "0 30 2 * * * *" }
}on_finish
Fires when a run of the same job the trigger is attached to finishes. It reacts only to that job's own runs, not to other jobs.
| Field | Type | Required | Default | Description |
|---|---|---|---|---|
delay | duration | no | none | How long to wait after a run finishes before starting the next. Omit it to start immediately. |
{
"kind": "on_finish",
"params": { "delay": "1h" }
}