Error 502 Bad Gateway on Nginx

[chiclean1]

I have a Wordpress blog running on Nginx.
Everything was working fine, but since from few days it started to return the error 502 Bad Gateway nginx/1.22.1

When I access the website on the browser I receive this error, then I wait some seconds and refresh the page and everything works.
I think that there is some issue during the “machine wake up”

Recents logs show:

2024-05-26 21:22:50.138 [416.847012] reboot: Restarting system
2024-05-26 21:22:50.138 WARN could not unmount /rootfs: EINVAL: Invalid argument
2024-05-26 21:22:49.386 ERROR error umounting /var/www/html: EBUSY: Device or resource busy, retrying in a bit
2024-05-26 21:22:48.634 ERROR error umounting /var/www/html: EBUSY: Device or resource busy, retrying in a bit
2024-05-26 21:22:47.881 ERROR error umounting /var/www/html: EBUSY: Device or resource busy, retrying in a bit
2024-05-26 21:22:47.138 2024/05/26 19:22:47 INFO SSH listening listen_address=[fdaa:7:f8d4:a7b:170:ced4:d456:2]:22 dns_server=[fdaa::3]:53
2024-05-26 21:22:47.131 [26-May-2024 19:22:47] NOTICE: exiting, bye-bye!
2024-05-26 21:22:47.131 [26-May-2024 19:22:47] NOTICE: Terminating ...
2024-05-26 21:22:47.130 WARN hallpass exited, pid: 328, status: signal: 15 (SIGTERM)
2024-05-26 21:22:47.130 ERROR error umounting /var/www/html: EBUSY: Device or resource busy, retrying in a bit
2024-05-26 21:22:47.129 INFO Umounting /dev/vdd from /var/www/html
2024-05-26 21:22:47.123 INFO Starting clean up.
2024-05-26 21:22:47.109 INFO Main child exited with signal (with signal 'SIGTERM', core dumped? false)
2024-05-26 21:22:46.226 INFO Sending signal SIGTERM to main child process w/ PID 327

[greg]

Hi,

Do you have your machines set to auto turn on/off, in the fly.toml ? That is the default and of course is cheaper (since you aren’t paying for an idle machine) but … it looks like it can cause issues. This post is about Laravel however the symptoms sound similar:

Laravel Octane 502 Bad Gateway When Machine Starts

Hey there! This does in fact appear to be an issue of Nginx sending a request to Octane before Octane has started. In fact, it’s sending the request before Octane is able to listen for requests, so increasing timeouts in Nginx doesn’t even work (Nginx just sees it as nothing being there, rather than sometime is listening but not responding in time). So, I’ve devised a fix that appears to work. It’s not my favorite fix, but I’m asking around to see if there’s something better (not just on X …

It might be worth (at least temporarily) keeping a minimum of 1 machine alive to confirm.

[chiclean1]

I added sleep 1 to my Dockerfile entrypoint script. Not beautiful but is working.