I am experimenting with cloud flows for the first time and I'm finding it a frustrating experience.
One of my frustrations is when I run a manual test on my flow. The trigger for my flow is "When an item is created or modified" on a SharePoint list. This is correctly configured to reach the list - I know that because it sometimes works (if incorrectly configured it would never work). But sometimes it doesn't. I click Test, the whirly wheel starts whirling and the prompt tells me to go and change an item in my list, I go and change an item and save it, and even create a new item, but the whirly wheel just keeps on whirling and eventually I get a timeout error.
The list is in a SharePoint online site (i.e. not on-premises), so it can't be to do with flaky network connections - it's all in the M365 infrastructure.
Is this just me or does anyone else observe such inconsistencies from Microsoft's infrastructure?
Thanks for the suggestion. I have just tried that, but to no avail - both by simply editing and making the change in SharePoint (the polling frequency on the connection is 5 minute - I've passed that now and nothing has happened) and by going back in and clicking the "Test" button. The flow checker gives no errors nor warnings - the flow simply does nothing. This is actually the most common behaviour - the times when it has run are very much in the minority, but significantly it is not never. I can't understand how Microsoft can have released something for so long that appears to give such intermittent performance. Why is nothing in this industry ever straightforward?
I have found that I need to back out of the editor screen for the flow to recognize changes in sharepoint.
Like you create a new flow and save and test. Use the left < arrow to go into the flow detail page or whatever you want to call it.
Then make a change in the SharePoint list.
It is like going to the main page establishes the links or something.