There's a problem with the flow's trigger - gateway error, error 400

[akharns]

On October 14th, I had a list (13) of flows fail.  All flows were monitoring on-premises directories for new files, over the gateway.  For all the flows, the log says "There’s a problem with the flow’s trigger.  Fix the trigger”.  By clicking on "fix the trigger", I can see error 400, "The requested action could not be completed. Check your request parameters to make sure the path <xxx> exists on your file system..."  When I edit the trigger step of the flow, browsing to the monitored folder gives the error "Error executing the api".  No messages were sent to me alerting me of the failure.  I discovered it accidentally, when I was expecting something that did not arrive.

I have had this problem before.  If I'm not mistaken, the instructions were to rebuild the trigger.  Since it is the trigger, I think this requires that I rebuild the entire flow.  Having to rebuild the entire suite of flows every time this error occurs is not reasonable.  

Another similar thread on this board has been open since July and no solutions have been provided.  I don't think this community has the attention of the engineers.

We are just starting to build this system of monitoring and reporting components, and if this behavior and recover process can be expected from time to time, I have to question whether the use of Flow is an acceptable solution.

Can someone reassure me that there is a way to resolve these issues?  Help, please!

[v-bacao-msft · Accepted solution]

Hi @akharns ,

Thank you for your feedback, and I am sorry that you have such an experience, but the help that the community can provide is limited, especially when you encounter problems that need to deal with On-Premise data source. It is difficult to solve this type simply by communicating in the forum.

The basic approach we take in dealing with similar issues in the forum is to try to reproduce similar issues. If it can be reproduced, then we will try our best to solve them and provide practical solutions to you.

If it can't be reproduced, we'll provide you with the correct steps to configure a similar Flow to see if it helps you. So the workaround you got before may be to reconfigure connection/Flow/trigger. We will do our best to try and solve your problem. But as mentioned earlier, the help provided in the forum is limited. Although we work hard, there are still some times when we can't do anything about it.

If this problem has a big impact on you, please consider contacting support for assistance. I believe that if communication is smooth, this problem can be solved very well.

Thanks again for your support of MS Flow.

Best Regards,

[akharns]

Thanks for your reply.  My sincere apologies if it sounded like I was chiding the support community.  I was really wondering if Microsoft has any involvement in seeing such pleas for help.  Obviously, the community of users can't correct problems with the product.  Thanks for directing me to support.

[akharns]

Btw, I will pursue Microsoft support, as suggested.

[akharns]

An update:   In a chat with microsoft support, they had me clear cache on Chrome.  Since this is a local action, I don't understand how it could sove a cloud-based process problem.  It resolves the gateway error I was seeing, with the "error executing the API" and the ability to browse to the target folder that was used in the trigger.  The "problem with the trigger" message is still there, even after closing it, but it seems that it clears after successfully executing.  I *think* I'm good on this.  

If anybody can explain why the Chrome cache could be relevant, I'd be interested in hearing.