Executing and Scheduling Workflows

Documentation Index

Fetch the complete documentation index at: https://trickest.com/llms.txt

Use this file to discover all available pages before exploring further.

​

Overview

Workflows can be run once on demand or on a recurring schedule. Both controls live in the top-right of the canvas in the Workflow Editor. For a one-off run, the editor offers three entry points:

• Execute: a one-click full run that re-executes every node.
• Smart Execute: an optimized run that reuses cached results from a previous run for nodes whose configuration and inputs have not changed.
• Advanced Execute: opens a configuration dialog where you can set fleet, parallelism, timeout, static IPs, and override workflow inputs before running.

For recurring runs, use scheduling to start a workflow at a fixed time and repeat it at a fixed interval.

---

​

Executing a Workflow

The execute control is the Execute button in the top-right of the canvas. The chevron next to it opens a dropdown with Smart Execute and Advanced….

Note Execution is disabled while a workflow has unsaved changes, while a save is in progress, while another run is starting, or while the workflow is scheduled. Save your changes (or cancel the schedule) before running.

​

Execute (Fresh Run)

Clicking Execute starts a fresh run that re-executes every node in the workflow, regardless of what previous runs produced. Use Execute when:

• You want a clean slate, for example before a comparison
• An external dependency changed (a remote source, a credential) that the optimizer cannot detect

The fleet and parallelism for the run come from the workflow’s saved defaults. To pick different values for one run, use Advanced Execute (see below).

​

Smart Execute

Smart Execute sits in the dropdown next to Execute. It re-executes only the nodes whose configuration or inputs changed since the last run. Nodes that are unchanged are served from cache (memoization), which is faster and cheaper than a full fresh run. Use Smart Execute when:

• You are iterating on a downstream node and do not want to re-run expensive upstream steps
• You expect most of the workflow to be unchanged from the previous run

The platform indicates how many nodes will execute and how many will be served from cache, for example: “Executing 3 of 7 nodes (4 cached)”.

Note Smart Execute needs a previous run to compare against. If there is no prior run, or if every node has changed, the run effectively becomes a fresh execution.

​

Advanced Execute

Advanced… sits below Smart Execute in the same dropdown. It opens a configuration dialog where you can override the defaults for this run. In the dialog you can configure:

• Fleet: which fleet the run executes on (Managed or a self-hosted fleet)
• Parallelism: maximum number of parallel jobs, up to the fleet’s available capacity
• Job Timeout Default: a global timeout applied to nodes that do not have a custom timeout
• Static IPs: enables static IP egress for the run (only available on Managed fleets when the feature is enabled)

To start the run with the chosen settings, click Execute at the bottom of the dialog.

Tip Use Advanced Execute the first time you run a new workflow to set the right fleet and parallelism, then enable Save as Default. Subsequent Execute and Smart Execute runs will use those defaults automatically.

​

Stopping an Active Run

While a run is active, Stop Run appears at the bottom of the Execute dropdown. Selecting it stops the run; nodes that are still in flight are wound down.

---

​

Scheduling Workflows

Scheduling automates execution at a specific time or at regular intervals. The scheduler is available from the top-right of the canvas, next to the Version history button. To open the scheduling dialog: In the dialog you configure when and how the workflow runs; the summary at the bottom updates as you change the options.

---

​

Scheduling Options

In the schedule dialog, you can configure the following:

• First Run - Choose the date and time for the first scheduled run. The first run must be in the future.
• Repeat - Choose how often the workflow runs as Every N minutes, hours, or days.
• Parallelism - Set maximum parallelism for your scheduled execution.
• Fleet - Choose the fleet where the workflow will execute.

As you update these values, the scheduling summary at the bottom of the dialog updates. To apply the schedule, click Schedule.

Note Scheduling a workflow makes it read-only. You must cancel the schedule before making edits.

---

​

Schedule Constraints

When scheduling workflows, keep the following constraints in mind:

• Workflow execution time vs interval
  The workflow execution time should be shorter than the repeat interval. If a workflow takes longer than the interval, subsequent runs will start based on the previous run’s completion.
• Fleet capacity and queuing
  Parallelism and fleet capacity must be available at runtime. If capacity is not available, the workflow will queue until resources are available.
• Workflow editing
  A scheduled workflow becomes read-only. To modify the workflow, cancel the schedule first.

---

​

Unscheduling Workflows

When a schedule is active, a Scheduled pill appears at the top of the canvas. The pill shows the repeat frequency (for example, Hourly, Every 2 days) and the next run date. The calendar icon in the top-right of the editor remains as an entry point to Edit Schedule, but cancelling is done from the pill. To cancel an active schedule: Once the schedule is cancelled, the pill disappears and the workflow becomes editable again.