1

I'm running Ubuntu 20.04.3 on a 1 GB RAM 1 VCPU 40GB SSD Lightsail instance. It runs NGINX serving a single Wordpress instance. The WP site is the web presence of our local Apple user group and gets very little traffic.

About once a month (plus or minus a few days) the Wordpress site goes away.

When that happens I'm unable to SSH into the Lightsail instance, neither through my SSH client (iTerm2) nor the AWS console. It appears to connect, but the screen is blank and there is no login prompt.

Rebooting the instance from the AWS console has no effect on the issue.

If I stop and restart the Lightsail instance everything comes back, but the public IP address has changed which means I need to make changes to DNS.

Syslog doesn't show anything I can see as the problem.

Any suggestions as to what I should be looking for?

Improve this question
4
  • Sounds to me like an IP addressing problem. Are you not using a static IP address?
    – Nmath
    Commented Jan 31, 2022 at 21:41
  • 2
    From the look of the AWS “Support” forum, this is a relatively common issue. There are a few possible solutions mentioned in the thread, but none are silver bullets. It may be better to switch to a self-managed EC2 instance 😕
    – matigo
    Commented Jan 31, 2022 at 23:15
  • @Nmath... yes it static addressing withing the limits imposed by AWS. Stopping and restarting the instance forces a new IP address.
    – NetFool
    Commented Feb 1, 2022 at 6:59
  • Qmatingo... I think your exactly right. I had almost reached that conclusion myself. I have an EC2 instance that's been running flawlessly for about 18 months. Now I just have to do a bit of web searching to see if there's an easy way to migrate from Lightsail to EC2. Given the view count on the forum thread you show above it's clear that nobody at Amazon gives a flying-fig about this problem.
    – NetFool
    Commented Feb 1, 2022 at 7:09

2 Answers 2

Reset to default
0

@matigo has the best solution. "Move it to EC2". Lightsail appears to be totally unsupported. I think, though, that I'll look at Linode and Digital Ocean VPS offerings first. No point rewarding AWS for bad service.

Improve this answer
1
  • I did move it to EC2 and the problem persisted. I added some additional logging and and it turns out that somewhere in the only two things running on the instance (nginx and wordpress+some plugins) there's a memory leak. When restarted it's using 64% of available memory and in a month it consumes the rest and crashes. A cron job that reboots the system once a week in the wee small hours of Sunday morning cured the problem.
    – NetFool
    Commented Jul 5, 2022 at 20:10
0

You might be using a low capacity instance in respect of website traffic. Lightsail instance crashes when heavy load is put on the system out of its capacity.

You can simply reboot the instance. It won't change your Server IP.

For permanent solution, Check your website traffic and chose instance as per your requirement.

Otherwise you can switch to EC2 with ELB or Autoscaling (Depends on your usage)

Improve this answer
1
  • 1
    When it crashes on Lightsail a simple reboot from the AWS Console won't work. You have to stop the instance and restart it, which gives you a new address. I switched to EC2 and the problem persisted ( see comment below for the real problem & solution ).
    – NetFool
    Commented Jul 5, 2022 at 20:14

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .