Latest stable is crashing

[MyWay]

Hi, I'm running eggdrop on arm, until 1.9.5 everything works fine, but when updating to latest stable, I get the following and then the container is crashing:

fetch http://dl-cdn.alpinelinux.org/alpine/edge/community/aarch64/APKINDEX.tar.gz
fetch https://dl-cdn.alpinelinux.org/alpine/v3.20/main/aarch64/APKINDEX.tar.gz
fetch https://dl-cdn.alpinelinux.org/alpine/v3.20/community/aarch64/APKINDEX.tar.gz
(1/4) Installing tzdata (2024b-r0)
(2/4) Installing tcl (8.6.14-r1)
(3/4) Installing tcl-lib (1.21-r1)
(4/4) Installing tcl-tls (1.7.22-r4)
Executing busybox-1.36.1-r29.trigger
OK: 33 MiB in 24 packages

Unfortunately the log isn't very helpful.

My docker compose:

services:
  eggdrop:
    image: eggdrop:1.9.5
    command:
      - /bin/sh
      - -c
      - |
        apk add --repository=http://dl-cdn.alpinelinux.org/alpine/edge/community tcl-lib tcl-tls && \
        exec /home/eggdrop/eggdrop/entrypoint.sh eggdrop.conf

[DArqueBishop]

I'm having the same issue. Running docker logs shows nothing; the container keeps crashing and restarting before anything is logged, apparently. The system I'm attempting to run the container on is an x86_64-based kvm VPS.

The following is my compose file:

services:
     eggdrop:
          image: eggdrop:stable
          container_name: eggdrop
          command: eggdrop.conf
          volumes:
               - "/var/local/eggdrop/data:/home/eggdrop/eggdrop/data"
               - "/var/local/eggdrop/contrib:/home/eggdrop/eggdrop/contrib"
               - "/var/local/eggdrop/logs:/home/eggdrop/eggdrop/logs"
               - "/var/local/eggdrop/netbots:/home/eggdrop/eggdrop/netbots"
          ports:
               - 4175:4175
               - 2010-2020:2010-2020
          restart: always
          stdin_open: true

[vanosg]

Thanks to both of you for reporting. Can anyone post anything further that covers the crash? I can't reproduce anything when I use the image pulled from docker hub. Appreciate any additional information you can provide, ie @MyWay what command you're running to get your error, or maybe @DArqueBishop more context on what is or isn't happening- maybe run it in console mode to see errors? Thanks

[DArqueBishop]

I ran docker compose up as opposed to docker compose up -d, and got this output before stopping with Ctrl-C:

root@athena:~/compose/eggdrop# docker compose up
[+] Running 2/2
 ✔ Network eggdrop_default  Created                                         0.2s
 ✔ Container eggdrop        Created                                         0.1s
Attaching to eggdrop
eggdrop exited with code 0
eggdrop exited with code 0
eggdrop exited with code 1
eggdrop exited with code 1
eggdrop exited with code 1
eggdrop exited with code 1
Gracefully stopping... (press Ctrl+C again to force)
[+] Stopping 1/1
 ✔ Container eggdrop  Stopped                                               0.0s

ETA: This is only happening with the stable branch. I switched back to 1.9 and the container is working properly.

[vanosg]

@DArqueBishop is this a new/fresh instance of eggdrop, or are you using an existing config file in the data directory?

[DArqueBishop]

It’s using an existing config file.

[vanosg]

@MyWay I would take a look at the example docker-compose file the docs provide at https://github.com/eggheads/eggdrop-docker/blob/master/docker-compose.yml and add in the required environmental variables (and other required settings), or mount an existing eggdrop configuration file.

@DArqueBishop I would double-check that the path and config file you are using exist where docker thinks it is- my working theory on this is the NICK and SERVER variables are not set in the docker-compose file, and there is no eggdrop config file it can find to load these variables from. For fun, you could try adding the environment variables into your compose file and see if that fixes it to narrow down the issue.

[MyWay]

@MyWay I would take a look at the example docker-compose file the docs provide at https://github.com/eggheads/eggdrop-docker/blob/master/docker-compose.yml and add in the required environmental variables (and other required settings), or mount an existing eggdrop configuration file.

@DArqueBishop I would double-check that the path and config file you are using exist where docker thinks it is- my working theory on this is the NICK and SERVER variables are not set in the docker-compose file, and there is no eggdrop config file it can find to load these variables from. For fun, you could try adding the environment variables into your compose file and see if that fixes it to narrow down the issue.

Keep in mind that it's working for both of us with the previous version. Any var break change? Any different setting for config file location? Because otherwise it can't be that.

[vanosg]

@MyWay If what you pasted is truly your full docker-compose file, then it should never have started with any version. What would be most helpful here is running it in foreground as I suggested above (no -d) and sharing the output.

[MyWay]

It was just to give you the command I used:

services:
  eggdrop:
    image: eggdrop:1.9.5
    container_name: eggdrop
    command:
      - /bin/sh
      - -c
      - |
        apk add --repository=http://dl-cdn.alpinelinux.org/alpine/edge/community tcl-lib tcl-tls && \
        exec /home/eggdrop/eggdrop/entrypoint.sh eggdrop.conf
    restart: always
    volumes:
      - ./data:/home/eggdrop/eggdrop/data
      - ./scripts2:/home/eggdrop/eggdrop/scripts2
      - ./Bot:/home/eggdrop/eggdrop/Bot
    env_file:
      - .env
    stdin_open: true
    tty: true

Change 1.9.5 to latest version to make it crash when starting. No error shown. Logs are those I pasted in the initial post.

@MyWay If what you pasted is truly your full docker-compose file, then it should never have started with any version. What would be most helpful here is running it in foreground as I suggested above (no -d) and sharing the output.

[vanosg]

@DArqueBishop thank you for your help testing the fix. I think (hope?) the most recent update just pushed to the eggdrop:develop image holds the fix. It takes about 24-48 hours for these changes to make it onto docker hub, so feel free to grab eggdrop:develop shortly and try it out.

The issue was some poor PID-checking logic that was added a version or two back that broke when 'set pidfile' wasn't present in the config file, as can be the case with a custom config file.

There's probably still room to break this though, so please let me know if you encounter any further issues (or if this fixes it for you). Thanks!