All updates

Retry a failed workflow run

Failed custom and data workflow runs can now be retried from their history, queuing a fresh run from the stored trigger while the original record stays put.

01 ·Retry a failed workflow run

A failed workflow run can now be retried from its history. Select one or more failed runs and choose Retry, and each is queued again, reusing the trigger data the original run captured so it starts from the same point without setting it up again. Runs that did not fail, or that belong to a workflow no longer deployed, are skipped, and the result tells you how many were queued and how many were skipped. The original failed run is left exactly as it was, and each retry is recorded as its own new run. It works for both Custom Workflows and Data Workflows, needs owner or admin access, and runs through the same queue as any other execution.

Failed run
Stored trigger
Starts from the same point
New run queued
Click Retry to run it
Original run stays unchanged · non-failed or undeployed runs are skipped
Illustration · A failed run is queued again from its stored trigger; the original stays put
Why it matters

A workflow that fails partway leaves a broker deciding whether to rebuild the trigger or let the work drop. Retrying from the stored trigger turns a failed run into a one-click redo, and keeping the original while creating a new run means the history still shows what happened both times, so an automation a broker relies on is easier to keep on track.

Also in this build
Fix

A further round of fixes for the activity notifications in the Channel Groups widget, covering replies, mentions, reactions, and direct messages.

15 Jul
Under the hood

Database connection handling was tuned so heavy automated traffic is easier to isolate as communities grow.

15 Jul