I am receiving multiple alerts (unhealthy host) from the cloud watch but my instance uptime is 15 days. All my applications are running fine on the server. But I want to find out why am receiving the alert.

Please help

  • What is the alert? – John Hanley Sep 14 '18 at 22:11
  • ALARM: "server V1-UnHealthyHosts-" in Asia Pacific. This is the error I got but server uptime is 17 days now. This instance is in under the ELB. – Linuxian Sep 17 '18 at 9:31
  • Uptime measures the uptime of the instance, not the services running within the instance. Enable CloudWatch and monitor items like CPU Credits (for Tx instances), Memory, CPU Utilization, etc. – John Hanley Sep 17 '18 at 19:18

What are the criteria for the alert? In more details, what is the threshold that triggers the alert? What is the interval between health checks?

AWS will generate an alert when the the number of unanswered health checks exceeds the threshold. Take the following example:

Assume you set health checks every 5 seconds, this means your load balancer will send a health check to the registered instance every 5 seconds. Now assume, that your alert is set up to send an alarm if the threshold exceeds or is equal to 1 every 5 minutes. As a result, as soon as the load balancer does not receive a response for any of the 60 health checks sent in these 5 minutes (which is probable due to network failure, or whatever reason), it will generate an alarm.

Check this link from the AWS CloudWatch Metrics

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.