2y ago

Upgrading from 0.17.2 to .18 failed. i need only the http server [Solved]

Hi.

i'm using ansible for installation and upgrading. My setup has an external nginx running which does the SSL part, so i want to remove the SSL and certbot parts of the /templates/nginx.conf file.

(i'm not sure if i have to edit the other nginx_internal.conf file) But with the below settings i get 'Bad gateway'.

Hoping someone can point me in the right direction. Thanks

limit_req_zone $binary_remote_addr zone={{domain}}_ratelimit:10m rate=1r/s;

# changelog
# 2023-06-27    Merge remote #5246 and remove certbot and ssl

server {
    listen 80;
    listen [::]:80;
    server_name {{domain}};

    location / {
      proxy_pass http://0.0.0.0:{{lemmy_port}};
      proxy_http_version 1.1;
      proxy_set_header Upgrade $http_upgrade;
      proxy_set_header Connection "upgrade";
      proxy_set_header X-Real-IP $remote_addr;
      proxy_set_header Host $host;
      proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
    }
}

access_log /var/log/nginx/access.log combined;

8 comments

Can you access it directly on http://0.0.0.0:{{lemmy_port}} with curl or something from the box itself?

502 is can't reach backend, so either nginx is failing to reach the backend or the backend is not running. Usually the backend's not running.
- Hi. thanks.
  
  i logged in and found the port number in the /etc/sites/sites_enabled/<domain>.conf
  
  i did the merge again, and via domain and the local port i get the following:
  
  i logged into portainer and found this message many times for <domain>_lemmy-ui_1:
  
  API error: FetchError: request to http://lemmy:8536/api/v3/site? failed, reason: getaddrinfo EAI_AGAIN lemmy
  
  in <domain>_lemmy_1 it says this many times:
  
  thread 'main' panicked at 'Failed to load settings file, see documentation (https://join-lemmy.org/docs/en/administration/configuration.html): LemmyError { message: None, inner: "unknown field `use_tls`, expected one of `smtp_server`, `smtp_login`, `smtp_password`, `smtp_from_address`, `tls_type`" near 40:12, context: SpanTrace [] }', crates/utils/src/settings/mod.rs:20:20 note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
  
  mm.. this use_tls field 🤔...
  
  On the ansible machine, i have this use_tls in the email section in inventory/host_vars/<domain>/config.hjson. i'll check if that field is still allowed
  
  ✅ in the defaults.hjson (mentioned in repy below), the use_tls isn't mentioned, so i removed that keyword, and Lemmy version 0.18 is up🙌
  
  Seems this documentation https://join-lemmy.org/docs/administration/configuration.html#full-config-with-default-values is a bit outdated
  
  it refers to {{#include ../../include/config/defaults.hjson}}
  
  but i couldn't find defaults.hjson except in the lemmy github here
  
  https://github.com/LemmyNet/lemmy/blob/main/config/defaults.hjson

8 comments