Choose the right trigger
The route is set up, the mapping is in place, a test run has gone through — started by hand. That leaves the question: When should it run by itself?
The triggers themselves are explained in Triggering. This page helps with the choice; how to set up the trigger is in the guide each section refers to.
Three things have a say before you choose
The data origin of the route. It is on the General tab, and it narrows the choice before you make one:
| Data origin | What is available |
|---|---|
| Dataset | schedule with Full or Incremental, event, by hand |
| Queue | schedule with the single run type Drain queue, by hand |
| Dispatch input | the route runs when the dispatching route hands it something |
The direction. An event trigger is only possible on a route that reads from SAP Business One. The selection dialog therefore offers Event only on export and internal routes — see React to a change in SAP.
The release. A trigger only starts a route once the switch in the Automation column on the Execution screen is switched on; until then, every schedule and every event rests. A newly set-up route that stands still usually has its reason exactly here.
Clock or event?
If both ways remain after the section above, where the rhythm comes from decides.
The clock. It fits when a fixed rhythm is enough — nightly, hourly, on Sundays — and when bundling is an advantage: a schedule collects what has changed since the last run and transfers it in one run. The way is in Run on a schedule.
The event. It fits when the change must be over there as soon as it is saved. For this, an event transfers individually: each change starts its own run with one job in it, and the run list fills up over the day. The way is in React to a change in SAP.
Both at once is allowed. A route can carry an event and, in addition, a nightly full run that catches up on what an event missed. The two complement each other.
The three questions that decide this in practice — how fast it must be over there, how much changes, whose rhythm applies — are in Triggering.
By hand always remains possible
Every route with its own source can be started, whether or not it has a trigger. For the first run of a new route, that is the usual way: first see what comes out, then set the clock. How a run is started by hand and what to choose there is in Push manually — a run or a single record.
When the question is really a different one
Two concerns sound like a trigger question but are solved differently.
"Route B should only run after route A." Two ways come close to that — see Put two routes in order.
"Too much goes out." Then it is about the what rather than the when. A trigger determines the rhythm; the scope is determined by the data filter of the route — see Narrow down with a data filter. If it is only about the first run, which should not push out the whole dataset at once, the way is in Limit the first run.
What comes next
Once the trigger is set up and released, the first run shows whether everything is in place: Check whether a run went through cleanly.
See also
- Routes — Automation (reference) — every field of the tab in detail
- Triggering (core concept) — the four ways to start a route, and how they work together