Load Balancer Issues

[Ben Jones]

Hi,

I'm trying to configure an HTTPS load balancer in GCE. I have this connected to 2 instance groups with one VM in each; these both work fine when accessed directly via their external IP address.

When I try and connect through the load balancer frontend IP, I get a 502 error with the message "The server encountered a temporary error and could not complete your request.". The logs show an associated failure:

jsonPayload: {

@type: "type.googleapis.com/google.cloud.loadbalancing.type.LoadBalancerLogEntry"

statusDetails: "failed_to_connect_to_backend"

}

I've read other discussions regarding this specific error which seem to point to a bad named port configuration on the instance groups but I can't see a problem with my configuration.

Is there anything else that this could be?

[Nicholas Elkaim]

Are your health checks coming back as healthy? If there are no instances marked as healthy then the LB won't be able to select a backend, causing an error. Keep in mind that only 200 responses will be accepted for health check probes, any redirect (302) planned as part of the response will mark the instance as unhealthy. Make sure your firewalls allow the probe ranges as well.

[Rasrizal Rosdi]

Maybe you need to check if there are unused instance groups with no endpoint inside the load balancer. Try to remove those if you didn't even use them.