After my 2 recent posts I didn’t want to give up on the failing Flows that I developed. In this post I will show you how to develop reliable Flows.
During this series I’m going through my conclusions as I try to get to developing reliable Flows. Feel free to skip straight though to the fixes in Part 3.
Posts in this series:
Part 1 – Twitter feeds turned into emails
Part 2 – SharePoint lists alerts with Flow
We saw in my first post about the reliability of Microsoft Flow the following flow:
Simply by doing what you would think should work and not really putting any effort into the Flow design a lot of the Flow instances were failing. So how do you fix this?
The answer is simple. Develop proper flows that check your data before using the data.
Then in advanced mode change the condition a little bit
Well this must work!
Nope, I tried quite q few different option, but it doesn’t seem to be possible to test for the triggerBody to be null.
Then I looked at the error message that I have been getting and tried the following:
And now we have success!
After a few days of running the Flow the flow analytics is showing only successful runs since the fixes were put in place.
Conclusions
Microsoft Flow seems to be very reliable when put under pressure as long as the appropriate error handling has been put in place. In my next post in this series I will dive further into the error handling option available in Microsoft Flow.