Logic App stops detecting Blob Storage changes

Question

We have a Logic App that monitors a blob storage path. On Sunday evening that blob storage location reached approximately 33k blob files. At this point, our logic app continued to trigger, once every minute, but each trigger was skipped as if it couldn't find any new files. After trying different things we began moving blobs out of the blob path. As soon as we did this the logic app began to detect all the missing files it had "skipped" since Sunday and process them. So, there seems to be some arbitrary limit that we can't seem to find documented anywhere where Logic Apps begin to fail to find new blob files.

We would really prefer not to have to move these files out of this location just so we can ensure the logic app continues to work. So, is there some arbitrary limit and if so what is it specifically, and what is the proposed best practice/workaround?

Answer 1

There is a Maximum trackable number of items inside of a virtual folder for trigger limit of 30k mention in the connector reference doc which I believe is what you are hitting.

The alternative is to leverage Event Grid and its corresponding trigger for Logic Apps, which gives you the added advantage of not having to poll the blob container and near instance trigger of the Logic App when a blob is added.