- Newest
- Most votes
- Most comments
Lightsail is very unreliable, I'm running two identical lightweight servers on LightSail and EC2, every few months the LightSail server will suddenly slow to a crawl that my website times out. I check the CPU utilization and it's under 1%, the disk usage is extremely low, credits are at a 100%. There's something wrong with the LightSail infrastructure, either some other VM from another user is taking up all the resources and starting the other VM's sharing the hardware or their setup is buggy. The only solution I've found to this is to stop and restart the instance and then it becomes a 100% responsive instantly. My guess is after a stop/start it spins it up on a different underlying hardware. I've never had an issue with the EC2 instance but the Lightsail instance (all of them) like clockwork will slow to a crawl and eventually fail (yup, the alarms go off because it stop responding to Amazon's own monitoring system) and I have to stop/start them again to fix it (or if I leave it for a 12+ hours or a few days, it'll eventually come back online but that's not a solution). Something is VERY VERY broken about Lightsail and Amazon refuses to acknowledge or fix it. I've been reporting this for about 3 years now.
This doc article may help with understanding metrics around CPU utilization, and when to move to a larger bundle size: https://lightsail.aws.amazon.com/ls/docs/en_us/articles/amazon-lightsail-viewing-instance-health-metrics#cpu-utilization-zones.
If you do decide to upgrade, this doc article provides details on how to do so: https://lightsail.aws.amazon.com/ls/docs/en_us/articles/how-to-create-larger-instance-from-snapshot-using-console
If you have any specific questions about your instance metrics or the instance upgrade process, I'd be happy to help.
Relevant content
- asked 5 years ago
- asked 2 years ago
- asked a year ago
AWS OFFICIALUpdated 10 months ago- AWS OFFICIALUpdated 3 months ago
- AWS OFFICIALUpdated 3 months ago
- AWS OFFICIALUpdated 6 months ago
I have not quite the same problem. The webserver just stops responding to requests, but I can ssh in and can see all the processes looking ok, and nothing in the logs, and no CPU usage or memory issues. A reboot brings the service back up. Happens randomly every few days on average.