woke up to monitoring alerts and this lovely page on every URL:
it's Apache proxying to Gunicorn, and it seems to happen only during traffic spikes. the error log mentions teh upstream a lot but i'm not suree if that's the cause or just a symptom. i want to fix this properly instead of adding a cron job that restarts services every hour. where do i start? i'll happily post whatever logs would help narrow this down. what am i doing wrong??
ahh right dmesg is full of Out of memory: Killed process lines. The VPS only has 1 GB of RAM.
yeahhh a 502 means teh proxy reacehd out to the app behind it and got a broken or no response. read the proxys error log first: sudo tail -n 100 /var/log/nginx/error.log for Nginx. connect() failed (111: Connection refused) means nothing is listening; upstream prematurely closed connection means teh app crashed or killed the request mid-way.
@ivan.jensen WOWWW the socket path changed after a PHP update and teh Nginx config still pointed at the old one. Updated fastcgi_pass, reloaded, done. Embarrassing, but fixed. seriously made my day 😭🙏
Sign in or create an account to reply.
