Lemmy Support @lemmy.ml sub_ubi @lemmy.ml 2y ago

lemmyBB: 502 Bad Gateway

I have a lemmy-ui instance working correctly, fresh ansible install. I haven't configured it yet, but it's up and running and test posts work fine.

I'm mostly interested in lemmyBB, so I installed that according to "Alongside existing Lemmy instance (native)" install instructions.

Everything seems to work fine during install, systemctl status lemmy_bb.service shows:

Jan 17 06:51:57 vps-f7b29755 lemmy_bb[484]: [2023-01-17T06:51:57Z INFO  rocket::launch_] (FileServer: assets) GET /assets/<path..> [10]
Jan 17 06:51:57 vps-f7b29755 lemmy_bb[484]: [2023-01-17T06:51:57Z INFO  rocket::launch] 📡 Fairings:
Jan 17 06:51:57 vps-f7b29755 lemmy_bb[484]: [2023-01-17T06:51:57Z INFO  rocket::launch_] Templating (ignite, liftoff)
Jan 17 06:51:57 vps-f7b29755 lemmy_bb[484]: [2023-01-17T06:51:57Z INFO  rocket::launch_] Shield (liftoff, response, singleton)
Jan 17 06:51:57 vps-f7b29755 lemmy_bb[484]: [2023-01-17T06:51:57Z INFO  rocket::launch_] Site data fetcher (request)
Jan 17 06:51:57 vps-f7b29755 lemmy_bb[484]: [2023-01-17T06:51:57Z INFO  rocket::shield::shield] 🛡️ Shield:
Jan 17 06:51:57 vps-f7b29755 lemmy_bb[484]: [2023-01-17T06:51:57Z INFO  rocket_dyn_templates::fairing] 📐 Templating:
Jan 17 06:51:57 vps-f7b29755 lemmy_bb[484]: [2023-01-17T06:51:57Z INFO  rocket::launch] 🚀 Rocket has launched from http://127.0.0.1:8703

When checking in with a browser, I get 502 Bad Gateway.

error.log

2023/01/17 07:05:13 [error] 528#528: *33 connect() failed (111: Connection refused) while connecting to upstream, client: 205.178.101.86, server: forums.domain.com, request: "GET / HTTP/2.0"$2023/01/17 07:05:13 [error] 528#528: *33 connect() failed (111: Connection refused) while connecting to upstream, client: 205.178.101.86, server: forums.domain.com, request: "GET /favicon.ic$2023/01/17 07:05:32 [error] 528#528: *50 connect() failed (111: Connection refused) while connecting to upstream, client: 205.178.101.86, server: forums.domain.com, request: "GET / HTTP/2.0"$2023/01/17 07:05:32 [error] 528#528: *50 connect() failed (111: Connection refused) while connecting to upstream, client: 205.178.101.86, server: forums.domain.com, request: "GET /favicon.ic$

You're viewing a single thread.

18 comments

This will be impossible to debug without the nginx (or whatever reverse proxy you are using) configuration. Because it looks like lemmybb is running but it is not being served properly.
- I'm using the config here: https://github.com/LemmyNet/lemmyBB/blob/main/docker/nginx.conf
  
  the only difference is that $lemmybb_domain is replaced with the bb ui domain name. Again, the normal lemmy ui works great, ansible install so everything including config is default.
  
  Is there something with the lemmy backend or UI I'm supposed to configure to let it know that lemmyBB exists?
  
  Rocket has launched from http://127.0.0.1:8703
  
  proxy_pass "http://0.0.0.0:8701";
  
  Make sure these ports are the same. Since your UI is running on 8703 you should use 8703 in the nginx config as well.
  
  I appreciate all the help! New error, 500
  
  systemctl status lemmy_bb.service
  
  Loaded: loaded (/etc/systemd/system/lemmy_bb.service; enabled; vendor preset: enabled) Active: active (running) since Tue 2023-01-17 17:55:48 UTC; 3min 42s ago Main PID: 20609 (lemmy_bb) Tasks: 2 (limit: 2318) Memory: 2.8M CGroup: /system.slice/lemmy_bb.service └─20609 /opt/lemmyBB/target/release/lemmy_bb Jan 17 17:58:24 vps-f7b29755 lemmy_bb[20609]: [2023-01-17T17:58:24Z INFO rocket::server] GET / text/html: Jan 17 17:58:24 vps-f7b29755 lemmy_bb[20609]: thread 'rocket-worker-thread' panicked at 'called `Option::unwrap()` on a `None` value', src/site_fairing.rs:62:53 Jan 17 17:58:24 vps-f7b29755 lemmy_bb[20609]: note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace Jan 17 17:58:24 vps-f7b29755 lemmy_bb[20609]: [2023-01-17T17:58:24Z ERROR _] Handler index panicked. Jan 17 17:58:24 vps-f7b29755 lemmy_bb[20609]: [2023-01-17T17:58:24Z WARN _] A panic is treated as an internal server error. Jan 17 17:58:24 vps-f7b29755 lemmy_bb[20609]: [2023-01-17T17:58:24Z WARN _] No 500 catcher registered. Using Rocket default. Jan 17 17:58:24 vps-f7b29755 lemmy_bb[20609]: [2023-01-17T17:58:24Z WARN lemmy_bb::site_fairing] error sending request for url (http://127.0.0.1:8536/api/v3/site): error trying to connect: tcp connect error: Connection refused (os error 111) Jan 17 17:58:24 vps-f7b29755 lemmy_bb[20609]: [2023-01-17T17:58:24Z INFO rocket::server] GET /favicon.ico image/avif: Jan 17 17:58:24 vps-f7b29755 lemmy_bb[20609]: [2023-01-17T17:58:24Z ERROR _] No matching routes for GET /favicon.ico image/avif. Jan 17 17:58:24 vps-f7b29755 lemmy_bb[20609]: [2023-01-17T17:58:24Z WARN _] No 404 catcher registered. Using Rocket default.```
  
  How are you running the lemmy backend? What port is it running on?
  
  Here's the config I use for the lemmy backend: https://github.com/LemmyNet/lemmy/blob/main/docker/prod/nginx.conf here's my lemmy.hjson https://github.com/LemmyNet/lemmy/blob/main/docker/prod/lemmy.hjson
  
  like I said, aside from domain name and required variables, everything is default, running a fresh vps. Ansible install for Lemmy, "install along other instance" instructions for lemmyBB.
  
  No messing with ports. Fresh debian 10, debian 11, ubuntu 22, all with the same result:
  
  Lemmy-UI works fine, lemmyBB returns a 502.
  
  Are you using docker at all? If yes, can you share the output of docker ps?
  
  Also, can you share your /etc/systemd/system/lemmy_bb.service file?
  
  Thank you for sticking with me!
  
  docker ps
  
  CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 88a017ba5903 dessalines/lemmy-ui:0.16.7 "docker-entrypoint.s…" 35 hours ago Up 22 minutes 127.0.0.1:8601->1234/tcp itsdomaincom_lemmy-ui_1 1299fcaeb678 dessalines/lemmy:0.16.7 "/app/lemmy" 35 hours ago Up 22 minutes 127.0.0.1:30850->8536/tcp itsdomaincom_lemmy_1 145c7d2f16f2 mwader/postfix-relay "/root/run" 35 hours ago Up 22 minutes 25/tcp itsdomaincom_postfix_1 737381a2939c asonix/pictrs:0.3.1 "/sbin/tini -- /usr/…" 35 hours ago Up 22 minutes 6669/tcp, 127.0.0.1:2112->8080/tcp itsdomaincom_pictrs_1 0f907e51bb9d postgres:12-alpine "docker-entrypoint.s…" 35 hours ago Up 22 minutes 5432/tcp itsdomaincom_postgres_1
  
  lemmy_bb.service
  
  [Unit] Description=lemmy_bb After=network.target [Service] Type=simple WorkingDirectory=/opt/lemmyBB/ Environment="LEMMYBB_BACKEND=http://127.0.0.1:8536" Environment="LEMMYBB_LISTEN_ADDRESS=127.0.0.1:8703" Environment="LEMMYBB_INCREASED_RATE_LIMIT=1" Environment="LD_PRELOAD=libjemalloc.so" ExecStart=/opt/lemmyBB/target/release/lemmy_bb Restart=always [Install] WantedBy=multi-user.target
  
  LEMMYBB_BACKEND=http://127.0.0.1:8536
  
  Try changing this to LEMMYBB_BACKEND=http://127.0.0.1:30850 and see if it works.
  
  New error, getting closer I think
  
  500: Internal Server Error The server encountered an internal error while processing this request. Rocket
  
  What do the lemmy_bb logs say?

18 comments