how to fix unhealthy pod issue via to startup probe

Question

hi
while trying to raise up my application on azure cluster AKS , i face issue with the pod healthy stage during the startup probe with the following error
Startup probe failed: Get "http://10.224.0.96:80/": dial tcp 10.224.0.96:80: connect: connection refusedtype: Warning

when i remove the startup probe from deployment yaml , the pod is become healthy officialy
the only log message that i have is the following :
[40m[32minfo[39m[22m[49m: Microsoft.Hosting.Lifetime[0]
Application is shutting down...

there is no more logs / log messages i dont know what is the rca of this issue

startupProbe:
              httpGet:
                path: /
                port: 80
                scheme: HTTP
              initialDelaySeconds: 300
              timeoutSeconds: 1
              periodSeconds: 15
              successThreshold: 1
              failureThreshold: 30

Answer 1

Hi @Dawod Kabha,

The kubelet uses startup probes to know when a container application has started, and when you removed it then you removed the validation of your application, the pod started but this does not mean your application is working properly, and from the logs, your application is not working and you need to increase the log level on your application to get more insight why the application is shutting down.