My Metrics Azure Monitoring show error http 5xx from some time and logs shows cup usage but monitor shows it was normal

Question

My azure monitor shows http 5xx error. It happens before but stable then it happen again from some days When I see diagnostic Detail logs it shows

Heartbeat took longer than "00:00:01" at "09/04/2021 01:08:21 +00:00"  

But when I show server CPU monitor log it shows normal usage you can see 30 days usage below

Here is Http error from azure monitor screen shot

Here is diagnostics reports shows why errors came

What I have tried is :
Restarted Server
Upscale instance Vertically/Horizontally

I am so confused what should I do ?

Answer 1

@Ahmad Mustafa , Firstly, apologies for the delayed response on this from over the weekend.

On which region have you provisioned your WebApp? App Service - Central US or Application Insights - West Europe? What’s the exact time in UTC that you experienced these issues?

You may review your Azure Service Health alerts, there were a couple of issues identified which were mitigated.

From Azure Portal alert:
Summary of Impact:
Between 15:44 UTC and 18:09 UTC on 03 Sep 2021, you have been identified as a customer using Apps Service in Central US who may have received intermittent HTTP 500-level response codes or timeouts when accessing their resources via HTTP or HTTPS.

Preliminary Root Cause:
An increase in the number of incoming requests to a single underlying endpoint led to that endpoint reaching an operational threshold, causing some connections to fail.
Mitigation: We manually throttled incoming traffic.

Next Steps: We will continue to investigate to establish the full root cause and identify necessary repairs and preventative actions to help deter issues like this from occurring.