Duplicating Playlists

[byurhannurula]

Hi, I have recently added Explo to Navidrome and first it had issues with the ListenBrainz API but then it started working but now every time it duplicates the weekly playlist and the daily jams looks like not running even

Here is my docker compose.yml

services:
  navidrome:
    image: deluan/navidrome:latest
    container_name: navidrome
    environment:
      - PUID=${PUID}
      - PGID=${PGID}
      - TZ=${TZ}
      - ND_SCANSCHEDULE=1h
      - ND_LOGLEVEL=info
      - ND_PASSWORDENCRYPTIONKEY=${ND_PASSWORDENCRYPTIONKEY}
      - ND_SESSIONTIMEOUT=24h
      - ND_LASTFM_APIKEY=${ND_LASTFM_APIKEY}
      - ND_LASTFM_SECRET=${ND_LASTFM_SECRET}
      - ND_LISTENBRAINZ_ENABLED=true
      - ND_AUTOIMPORTPLAYLISTS=true
      # - ND_PLAYLISTSPATH=/music/playlists
      - ND_BASEURL=/ # Important for reverse proxy
      - ND_ENABLETRANSCODINGCONFIG=true
      - ND_ENABLEDOWNLOADS=true
    volumes:
      - ./navidrome/data:/data
      - ./music:/music:ro
    ports:
      - 4533:4533
    restart: unless-stopped
    networks:
      - default
  explo:
    image: ghcr.io/lumepart/explo:latest
    container_name: explo
    volumes:
      - ./.env:/opt/explo/.env
      - ./music/explo:/data/
    environment:
      - TZ=${TZ}
      # Runs weekly, every Tuesday 15 minutes past midnight
      - WEEKLY_EXPLORATION_SCHEDULE=15 00 * * 2
      # Run weekly exploration with default settings
      - WEEKLY_EXPLORATION_FLAGS=
      # Uncomment _SCHEDULE and _FLAGS variables to enable fetching different playlist
      #- WEEKLY_JAMS_SCHEDULE=30 00 * * 1 # Runs weekly, every Monday 30 minutes past midnight
      #- WEEKLY_JAMS_FLAGS=--playlist=weekly-jams --download-mode=skip # Get tracks from weekly-jams, and only add tracks that are found locally to playlist

      - DAILY_JAMS_SCHEDULE=15 01 * * * # Runs daily, every day 15 minutes past 1PM
      #- DAILY_JAMS_FLAGS=--playlist=daily-jams --download-mode=skip # Get tracks from daily-jams, and only add tracks that are found locally to playlist

      - EXECUTE_ON_START=true
      #- START_FLAGS=--persist=false # if needed, set startup flags for testing
    depends_on:
      - navidrome
    restart: unless-stopped
    networks:
      - default

networks:
  default:
    driver: bridge

TZ=Europe/.......
PUID=1000
PGID=1000

# Navidrome
ND_USERNAME=.............  
ND_PASSWORD=........................

ND_PASSWORDENCRYPTIONKEY=...................

ND_LASTFM_APIKEY=................
ND_LASTFM_SECRET=.............

# ListenBrainz
LISTENBRAINZ_USER=kilogram4007

# Explo
EXPLO_SYSTEM=subsonic
SYSTEM_URL=http://navidrome:4533
SYSTEM_USERNAME=................
SYSTEM_PASSWORD=...................
USE_SUBDIRECTORY=true
DOWNLOAD_SERVICES=youtube
YOUTUBE_API_KEY=.....................
SINGLE_ARTIST=true
PERSIST=true
LOG_LEVEL=DEBUG

And here are logs when fetches the weekly playlist: https://pastebin.com/nV562Ksb
And logs on restart the stack: https://pastebin.com/tTSL1ns5

And as a result here is what I got after some testing (Week 50 was also duplicated I just deleted one)

[LumePart]

Hey!
It looks like you have DAILY_JAMS_FLAGS commented, you need to uncomment it in docker-compose, otherwise it runs with default flags (fetches weekly-exploration playlist)

And I suggest commenting EXECUTE_ON_START, that also duplicates playlist if you also have it running on a schedule

[byurhannurula]

EXECUTE_ON_START was just for debugging in order to not wait for crons

Oh I thought FLAGS are optional params so I left it commented how I got it from example.env now I uncommented it and made it empty like the WEEKLY one and changed scheduler to run few mins after deployin the updated compose file but still it did not triggered the cron

- DAILY_JAMS_SCHEDULE=42 13 * * * # Runs daily, every day 15 minutes past 1PM
- DAILY_JAMS_FLAGS=

[LumePart]

Do the container logs show that Explo registered the job with the correct timestamp?

Also, leaving the DAILY_JAMS_FLAGS empty will import a weekly-exploration playlist, you have to specify what playlist you want with the --playlist flag, your full docker-compose had the correct recommended flags set (just commented out)