[–] 1 point 1 year ago

on mine I'm running forgejo, searxng, navidrome, vaultwarden, wireguard, jellyfin, RomM, some VM stuff mainly for compiling, and a bunch of other small stuff like glance.

Then I have a "junk drawer" on it where I just play around with random things in dockers and generally forget about them.

  • source
  • [–] 1 point 1 year ago

    I run nearly all my Docker workloads with their data just in the home directory of the VM (or LXC actually since that's how I roll) I'm running them in, but a few have data on my separate NAS via and NFS share - so through a switch etc with no problems - just slowish.

  • source
  • parent
  • context
  • [–] 1 point 1 year ago*

    @RareBird15 @selfhost @selfhosted @selfhosting Inspired by you I've locally hosted Joplin server, partly to migrate off ColorNote. It seems great for moving text/notes between my devices while keeping a core copy in my NAS. Thanks.

  • source
  • [–] [S] 1 point 1 year ago (1 child)

    Interestingly, I did something similar with Linkwarden where I installed the datasets in /home/user/linkwarden/data. The dam thing caused my VM to run out of space because it started downloading pages for the 4000 bookmarks I had. It went into crisis mode so I stopped it. I then created a dataset on my Truenas Scale machine and NFS exported to the VM on the same server. I simply cp -R to the new NFS mountpoint, edited the yml file with the new paths and voila! It seems to be working. I know that some docker container don't like working off NFS share so we'll see. I wonder ho well this will work when the VM is on a different machine as the there is a network cable, a switch, etc. in between. If for any reason the nas goes down, the docker containers on the Proxmox VM will be crying as they'll lose the link to their volumes? Can anything be done about this? I guess it can never be as risilient as having VM and has on the same machine.

  • source
  • parent
  • context
  • [–] 1 point 1 year ago (2 children)

    Great. There's two volumes there - firefly_iii_upload & firefly_iii_db.

    You'll definitely want to docker compose down first (to ensure the database is not being updated), then:

    docker run --rm \
      -v firefly_iii_db:/from \
      -v $(pwd):/to \
      alpine sh -c "cd /from && tar cf /to/firefly_iii_db.tar ."
    

    and

    docker run --rm \
      -v firefly_iii_upload:/from \
      -v $(pwd):/to \
      alpine sh -c "cd /from && tar cf /to/firefly_iii_upload.tar ."
    

    Then copy those two .tar files to the new VM. Then create the new empty volumes with:

    docker volume create firefly_iii_db
    docker volume create firefly_iii_upload
    

    And untar your data into the volumes:

    docker run --rm \
      -v firefly_iii_db:/to \
      -v $(pwd):/from \
      alpine sh -c "cd /to && tar xf /from/firefly_iii_db.tar"
    
    docker run --rm \
      -v firefly_iii_upload:/to \
      -v $(pwd):/from \
      alpine sh -c "cd /to && tar xf /from/firefly_iii_upload.tar"
    

    Then make sure you've manually brought over the compose file and those two .env files, and you should be able to docker compose up and be in business again. Good choice with Proxmox in my opinion.

  • source
  • parent
  • context
  • [–] [S] 3 points 1 year ago

    The first rule of containers is that you do not store any data in containers.

    Do you mean they should be bind mounts? From here, a bind mount should look like this:

    version: '3.8'

    services: my_container: image: my_image:latest volumes: - /path/on/host:/path/in/container

    So referring to my Firefly compose above, then I shoudl simply be able to copy over the /var/www/html/storage/upload for the main app data and the database stored in here /var/lib/mysql can just be copied over? but then why does my local folder not have any strorage/upload folders?

    user@vm101:/var/www/html$ ls index.html

  • source
  • parent
  • context
  • [–] [S] 1 point 1 year ago (3 children)

    Here is my docker compose file below. I think I used the standard file that the developer ships, simply because I was keen to get firefly going without fully understanding the complexity of docker storage in volumes.

     The Firefly III Data Importer will ask you for the Firefly III URL and a "Client ID".
    # You can generate the Client ID at http://localhost/profile (after registering)
    # The Firefly III URL is: http://app:8080/
    #
    # Other URL's will give 500 | Server Error
    #
    
    services:
      app:
        image: fireflyiii/core:latest
        hostname: app
        container_name: firefly_iii_core
        networks:
          - firefly_iii
        restart: always
        volumes:
          - firefly_iii_upload:/var/www/html/storage/upload
        env_file: .env
        ports:
          - '84:8080'
        depends_on:
          - db
      db:
        image: mariadb:lts
        hostname: db
        container_name: firefly_iii_db
        networks:
          - firefly_iii
        restart: always
        env_file: .db.env
        volumes:
          - firefly_iii_db:/var/lib/mysql
    
      importer:
        image: fireflyiii/data-importer:latest
        hostname: importer
        restart: always
        container_name: firefly_iii_importer
        networks:
          - firefly_iii
        ports:
          - '81:8080'
        depends_on:
          - app
        env_file: .importer.env
    
      cron:
        #
        # To make this work, set STATIC_CRON_TOKEN in your .env file or as an environment variable and replace REPLACEME below
        # The STATIC_CRON_TOKEN must be *exactly* 32 characters long
        #
        image: alpine
        container_name: firefly_iii_cron
        restart: always
        command: sh -c "echo \"0 3 * * * wget -qO- http://app:8080/api/v1/cron/XTrhfJh9crQGfGst0OxoU7BCRD9VepYb;echo/" | crontab - && crond -f -L /dev/stdout"
        networks:
          - firefly_iii
    volumes:
       firefly_iii_upload:
       firefly_iii_db:
    
    networks:
      firefly_iii:
        driver: bridge
    
  • source
  • parent
  • context
  • Whatever you do, make sure you have working backups first.

    I imagine you could copy the docker volumes over, but that's more work than of they're "mounts", in which case you can just copy the corresponding on the host. Use scp or rclone or whatever to copy the files over

  • source
  • [–] 2 points 1 year ago

    Something that I noticed people seem to not realize with trash guides is that you can always customize it a bit.

    I disabled remux on mine sonarr/radarr and I also set a slightly lower max file size for encodes and webdl. Still grabs the good quality stuff to me based on the curated groups from trash guides.

    You can play with the settings of your setup you don't have to follow the guide one to one

  • source
  • parent
  • context
  • [–] 2 points 1 year ago (4 children)

    I'm not clear from your question, but I'm guessing you're talking about data stored in Docker volumes? (if they are bind mounts you're all good - you can just copy it). The compose files I found online for FireflyIII use volumes, but Hammond looked like bind mounts. If you're not sure, post your compose files here with the secrets redacted.

    To move data out of a Docker volume, a common way is to mount the volume into a temporary container to copy it out. Something like:

    docker run --rm \
      -v myvolume:/from \
      -v $(pwd):/to \
      alpine sh -c "cd /from && tar cf /to/myvolume.tar ."
    

    Then on the machine you're moving to, create the new empty Docker volume and do the temporary copy back in:

    docker volume create myvolume
    docker run --rm \
      -v myvolume:/to \
      -v $(pwd):/from \
      alpine sh -c "cd /to && tar xf /from/myvolume.tar"
    

    Or, even better, just untar it into a data directory under your compose file and bind mount it so you don't have this problem in future. Perhaps there's some reason why Docker volumes are good, but I'm not sure what it is.

  • source
  • [–] 5 points 1 year ago (1 child)

    The first rule of containers is that you do not store any data in containers.

    The second rule of containers is that you run them from a versioned config with proper volumes and tagging. Always.

    If you obey these rules, then it's as simple as moving the volumes to another host and starting your containers. They're fully portable that way.

  • source
  • [–] 3 points 1 year ago

    Assuming truenas does NFS, I would mount the old docker data into the new docker VM. Stop all running containers and copy the data, copy so that you have a backup should you need to revert.

    Make sure all containers are in compose files, up date them for new data location.

  • source
  • [–] 4 points 1 year ago (1 child)

    I think the real question is going to be whether your power bill is going to be worth it. Running a large chassis server to power such old and small drives seems like kind of a bad trade considering how much more sense storage is now.

    A RAID5 array with these will yield you 28TB usable, and then you'd have to worry about getting the same drive dimensions to replace one if it goes down in the coming years, which is going to be tricky.

  • source
  • If you've been trying for a while and still haven't gotten it working, then I'm probably missing something regarding your requirements.

    But at first glance, it seems like it would be solved with 2 server blocks, 1 for 80, 1 for 443, each with their own proxy config passing / to their respective target ports.

  • source
  • parent
  • context
  • view more: ‹ prev next ›