- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
‎03-08-2023 04:43 AM
Hi All,
We are developing an automation via SNOW that triggers an external automation engine to perform an action. When the action is performed depends on a number of conditions and may be days away. The action is triggered in a workflow. What would the best practice be to have the workflow wait for the action to complete? We use timer and turnstile for shorter actions (to perform within 24 hours), but I would prefer to avoid polling and do wait for condition or event instead.
Solved! Go to Solution.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
‎03-08-2023 05:19 AM - edited ‎03-08-2023 05:20 AM
When you "wait", you keep that workflow hanging, consuming system resources, which can degrade the performance of your instance. This is not the case when you just end the workflow at a certain point, and get another one kicked off when it is needed.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
‎03-08-2023 04:53 AM
You should branch off whatever you have after the "wait" condition into a separate flow or workflow or other business logic, with it's own trigger, whereas the trigger could be an inbound REST call (or webhook) for instance from the 3rd party system. So in a nutshell, 2 (or more) workflows instead of doing it in one.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
‎03-08-2023 05:03 AM
How does having two separate workflows help here as opposed to just having a 'Wait for event" or "Wait for condition", and continuing after the event or condition occur?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
‎03-08-2023 05:19 AM - edited ‎03-08-2023 05:20 AM
When you "wait", you keep that workflow hanging, consuming system resources, which can degrade the performance of your instance. This is not the case when you just end the workflow at a certain point, and get another one kicked off when it is needed.