Load Balancer Backend Memebrs Health Status

Mohamad Swead 0 Reputation points
2023-07-25T10:04:26.1366667+00:00

Hello,

I have a standard Azure LB configured with two servers in the backend. TCP health prob is configured to monitor the backend servers over port 80 with an interval of 5 seconds.

When shutting down one of the VMs in the backend, the status of the stopped VM remains healthy for over than 5 minutes before being marked as unhealthy. Almost the same time is needed to re mark the VM as healthy after starting the VM.

Any idea why having this unexpected behavior ?

Thanks in advance.

Azure Load Balancer
Azure Load Balancer
An Azure service that delivers high availability and network performance to applications.
421 questions
{count} votes

1 answer

Sort by: Most helpful
  1. msrini-MSFT 9,271 Reputation points Microsoft Employee
    2023-07-28T03:39:05.9833333+00:00

    Hi, I believe there are no data path issues when you turn off one for the backend VM. If that is the case, then load balancer is working as expected. It's the metric which shows unhealthy in a delay which I think creates confusion here. If you can repeat the same excercise again and bring back the VM up after 15 mins. Once the VM is back online, can you please take a screenshot of the probe metric and share it here? Ideally I should see the 15 mins probe down in the metrics.

    0 comments No comments