Ghost crashes out every few hours and won't restart

If you’re looking for some help, it’s important to provide as much context as possible so that people are able to assist you. Try to always mention:

  • What’s your URL? This is the easiest way for others to debug your issue
  • What version of Ghost are you using? 2.29.0
  • What configuration? Ubuntu 18.4 LTS running on Google Cloud VM
  • What browser? Chrome
  • What errors or information do you see in the console? None
  • What steps could someone else take to reproduce the issue you’re having?

I installed Ghost on a Google Cloud VM a week or so ago, and I’ve been struggling with a crash issue I can’t figure out. Every few hours, Ghost will go down and I’ll see an NGINX 502 error when going to the URL. I looked in the logs (both main and the error) in contents/logs, and there are no clues in there what is happening. To fix it, I have to manually do a ghost start (works most of the time) or shut the server down and start it again.

Any ideas what could cause this or how I can better try to find the culprit?

Hi @Alex_K, when you setup the Ghost site did you follow these instructions?

What does it say in /var/log/nginx/error.log ?
What does ‘sudo systemctl status nginx’ say? (Assuming Ubuntu18 uses systemd)

Hi guys,

Thanks for your replies!

@DavidDarnes we setup the server as compute engine VM (rather than app engine), so we were able to follow the Ubuntu setup guide. We did think about going the App Engine route, but the tutorial wants us to use the NPM module version of ghost, and it looks like the module is no longer supported?

@jeff The nginx error log had a few entries showing “connection refused while connecting upstream”. There was also an SSL_do_handshake() error which further said “tls_early_post_process_client_helllo:version too low”

The nginx status is active and running since I started the server.

Let me know if any of this sparks an AHA for you guys!