Cannot ping AWS Lightsail Server, even though it says it's running. Ideas?

Hello,

Thanks for using Lightsail!

Lightsail instances can slow down and appear as unresponsive when the instance consistently consumes more CPU than the Lightsail plan provides as a baseline. Lightsail uses burstable instance types that allow you to consume more CPU than your baseline for short periods of time. However, if an application continues to use more CPU beyond the plan's resources for long periods of time, the performance of the instance may suffer. This can happen for large or CPU-intensive applications that run on a small plan size or for applications experiencing consistently high traffic on small plan sizes. We are working on ways to make it easier to diagnose when this issue is happening.

Here are several options to address this issue:

1. Scale up your instance
   If you find your instances are encountering this scenario, you can scale up your instance to the next plan size for improved performance. You can increase your plan size by taking a snapshot of your instance and then recreating a new, larger instance from your snapshot:
   https://lightsail.aws.amazon.com/ls/docs/en/articles/how-to-create-larger-instance-from-snapshot-using-console

2. Disable unnecessary, resource-consuming code
   You can try to pinpoint resource-intensive processes that are not essential to your application, but are consuming your CPU and disable them. For example, in applications like WordPress, custom themes, WordPress plugins, or other processes may be unnecessarily consuming your CPU in the background. If these element are not essential, you can turn them off.

3. Use multiple Lightsail Instances
   Another way to scale out your application is by using multiple instances to distribute the load. This is particularly useful if you are experiencing high CPU utilization due to traffic to your application or website. Lightsail load balancers are designed to make scaling your application easy. See more in Lightsail docs: https://lightsail.aws.amazon.com/ls/docs/en/articles/understanding-lightsail-load-balancers

4. Move your application to EC2
   The largest Lightsail plan can support some of the most demanding applications. However, if you find you need more CPU or memory than Lightsail instances provide, we recommend you use some of the more powerful EC2 instances like M5 or C5: https://aws.amazon.com/ec2/

Finally, if you need help troubleshooting this issue, please open a support ticket. Please note that information about the applications and processes running on your instance are not visible to the Lightsail team or the support team, since customers own all content on their instance. The more information you provide about your application, the easier it will be for us to help you determine the best way to resolve the issue. I hope this helps!

Best,
Sowjanya

Thank you for the detailed response. The current CPU usage is 3.98%. I cannot see the metrics from previous days to discover if it was an issue 3 days ago when the website went down. It's holding 2 small Wordpress websites for a small business that is not using any digital marketing tactics to drive activity at the moment. Yes, we intend to purchase the $29 tech support fee today.

The burst capacity is 100%. Question: Do we lose "credits" of burst capacity after rebooting the server?

Edited by: yayaAWS on Oct 2, 2020 5:55 AM

Hello, I am updating this thread to share what is happening so far. On October 2nd, I upgraded to developer status for tech support. It's now October 6th and I have received no response. The only change is that the ticket has gone from "unassigned" to "work in progress". As a new customer (signed up 3 months ago), we were checking to see if upgraded tech support was a viable option in this case. In 20+ years, I've never experienced a website down this long.

Maybe a dumb question but are pings turned on, on your instance's "networking" tab?
I believe they are turned off by default. I had to turn mine on.