Currently, we have a total of 10 unhealthy hosts under a specific ELB, hosts are in different AZs.
aws_elb_un_healthy_host_count_average metric returns 10 unhealthy hosts for each AZ: aws_elb_un_healthy_host_count_average{load_balancer_name="foo",availability_zone="us-west-2c",} 10.0 aws_elb_un_healthy_host_count_average{load_balancer_name="foo",availability_zone="us-west-2b",} 10.0 aws_elb_un_healthy_host_count_average{load_balancer_name="foo",availability_zone="us-west-2a",} 10.0 therefore, sum by (load_balancer_name) (aws_elb_un_healthy_host_count_average{load_balancer_name="foo"}) returns 30 unhealthy hosts, which is incorrect (should be 10) On the other hand, aws_elb_healthy_host_count_average returns just one value, without availability_zone label, and therefore returns a correct number. Using cloudwatch_exporter-0.8.0 image -- You received this message because you are subscribed to the Google Groups "Prometheus Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to prometheus-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/27564926-cc9a-4b56-9fcc-014a1dfdb58en%40googlegroups.com.