On schedule
Start a flow on a cron schedule, optionally spread over partition keys.
Starts a flow on a cron schedule.
concurrency decides what happens when a run is still going at the next fire — skip it, queue it, or replace it. spread staggers one logical fire across partition keys over overMinutes, so a thousand environments do not all start at 02:00; each partition arrives as its own run with its key on partitionKey. catchUp replays fires missed while the application was paused.
At a glance
Module type
OnSchedule
Category
Events
Default tier
durable
Runs on
durable
Starts a flow
yes
Holds a body
no
Keeps state between runs
no
Available in
every application type
Parameters
cron
string · required · defaults to "0 2 * * *"
The schedule, as a cron expression.
concurrency
"replace" | "skip" | "queue" · required · defaults to "skip"
What to do when the previous run is still going: skip this fire, queue it, or replace the run.
catchUp
boolean · required · defaults to false
Replay fires that were missed while the application was paused.
timezone
string | null · optional · defaults to null
Null means the organisation timezone.
spread
object · optional · defaults to null
Stagger one fire across partition keys, so a large fleet does not all start at once. Give it partitionBy — a list of keys, or a dataset column to read them from — and overMinutes to spread across.
Inputs
None.
Outputs
event
event
event, json
The event envelope: {kind, at, payload}.
firedAt
scalar<datetime>
scalar, json, text, signal
When this run fired. The envelope's at, stamped as the run arrives, so a spread or a caught-up fire carries the instant it really ran and not the nominal cron time.
partitionKey
scalar<string>
scalar, json, text, signal
The partition this run covers, when spread is set.
Last updated