Azure App Service unavailable to other App Service while clearly available from outside

John K 6 Reputation points
2020-07-02T09:13:11.657+00:00

I'm don't think has been asked before, at least none of the suggested questions really matched my problem.

I have two app services sitting in the west-europe data center (Holland), lets call them Webapp and Microservice. Webapp calls Microservice via it´s public host name (no virtual networks involved at all). Webapp also has a health check that makes an HTTP request to Microservice in order to be considered healthy.

Some times, quite often actually, I get an alert from our monitoring system that Webapp is unhealthy due to the request to Microservice failing (timeout). What I usually do then is to fire off a request from my local computer to Microservice, using the same address and port, just to verify this, and most of the times I can actually access it. So we have a situation where I can successfully make a request to Microservice from my local computer in Sweden, but Webapp, sitting in the same data center in Holland, cannot reach it.

This happens intermittently. It can go weeks without any incident, and then there will be a period where it comes and goes, like today where it's been down multiple times.

Is there any obvious reason why this might be?

Azure App Service
Azure App Service
Azure App Service is a service used to create and deploy scalable, mission-critical web apps.
7,334 questions
0 comments No comments
{count} votes

1 answer

Sort by: Most helpful
  1. Ryan Hill 26,951 Reputation points Microsoft Employee
    2020-07-06T20:48:37.48+00:00

    Hi @JohnK-7484,

    Apologies for the late reply. Are you using Health Check preview from your web app set to your Microservice endpoint? Are there any logs you can share from the Diagnose & Solve problems blade? Does the interval between timeouts occur on a interval or at random? One thing I can think of which you didn't mention was if you app service hosting your Microservice was set to 'Always On' or not. If it isn't, then the health ping could be "waking up" your Microservice and there is startup code that causing the timeout delay.