We have developed several Flows which work mostly with SharePoint API. Usually we use "Get item", "Update item" and "Send an HTTP request to SharePoint" actions.
Most of the time everything works as it should. However, sometimes actions fail seemingly because SharePoint Online server becomes unavailable:
Message in that JSON shows this:
I've seen this "HTTP Error 503. The service is unavailable." error message on-prem. Usually this happens during IISRESET. And I do understand Microsoft has to restart SharePoint Online servers as well, however, Flow should be able to handle this kind of problems.
As far as I can tell, Flow automatically retries request if it receives HTTP status 408 (request timeout), 429 (too many requests) or anything from 5xx. In this case SharePoint return message clearly states "HTTP Error 503". However, action itself displays "Status code: 400". I think this is the reason why Flow fails - it does not even try to resend request to the same URL.
Currently I don't have any viable solution. I could add "run after failed" path to each SharePoint action and put them in some kind of loop, but this seems very complicated.
Has anyone experienced this problem? Maybe there is a normal solution which does not involve redoing everything from the start?
