Health probe says the endpoints (Azure Container Instances) are degraded

1
votes

I have set up a health probe to ping Azure Containers from my load balancer so that requests are forwarded to only healthy nodes. However, I am getting "Degraded" status despite the containers being up and running. I am aware this has got to do with the reponse the health probe gets from the IP Address but I cannot figure out what changes do I need to make to my container settings to ensure that it works as expected

1
azureazure-container-instances
Do you have more questions in my reply? If the reply is helpful or others can benefit from it, you may accept it like this. – Nancy Xiong

1 Answers

0
votes

For Azure load balancer backend pool management, there are two ways of configuring a backend pool:

Network Interface Card (NIC)
Combination of IP address and Virtual Network (VNET) Resource ID

You could deploy your ACI into a VNet then add the IP address in the backend pool, then you can deploy load balancer rules and health probe for your backend ports. You could configure the appropriate health probe type for your backends. Read Load Balancer health probes

For example, I have an application that exposes port 8000 within the container in a VNet.

Load balancer rule enter image description here Health probe enter image description here