ootri

Members
  • Posts

    4
  • Joined

  • Last visited

ootri's Achievements

Noob

Noob (1/14)

1

Reputation

  1. I think this is happening because the docker image is running as user id 99, which doesn't actually exist and this non-existent user doesn't have access to chmod so it can't change that directory. This can be fixed by switching to "Advanced View" and putting the following in the "Extra Parameters" field: --user 65534:65534 Then in the Unraid command line, set matching permissions to your share: chown -R 65534 /mnt/user/backup/github # Or wherever you're putting the files I have a suspicion this approach might break with container updates and it will certainly break if/when your permissions get reset, so as an alternative to the above, you can simply put in the "Extra Parameters" field: --user 0:0 This will run the docker as root which we should all know by now isn't ideal, but it is the default for docker (you probably have others that do this too like mariadb and redis) and this shouldn't be a problem because docker provides isolation between the container and host, but make sure you do not run this as "privileged".
  2. There really isn't anything special about exposing a vm port, just open the port in the windows firewall. With that said, there's a lot I can talk about with what you're doing because I had to do something similar, here's how I did it: There exists a RabbitMQ docker image in the apps store, I had just posted about it a few days ago here. Running the docker image would be much more resource friendly than a vm, especially a windows vm. You should have two instances of RabbitMQ, one in each network. RabbitMQ has a great plugin called shovel which pushes messages from a source to destination. It's easy to use and highly reliable. I had tons of issues getting messages across my VPN link for various reasons, but all those issues went away with an extra RabbitMQ instance and the shovel plugin. To add the shovel plugin (and shovel management) to the docker images, just add to the "enabled_plugins" file: rabbitmq_shovel,rabbitmq_shovel_management Regardless if you use a vm or docker and regardless if you have one RabbitMQ instance or two, you need a secure and fast way to connect the networks. The obvious choice is WireGuard, but I suggest you also look into Tailscale. There's also a docker image for Tailscale which is super easy to use. If you continue with the windows vm path, just be sure to open the ports in windows firewall.
  3. Thank you for making this template. Exactly what I was searching for. A few notes on this template. The 3 environment variables in the template need to be removed. Reviewing the docs, it seems there is some confusion with how the examples translate to unraid. "hostname" should be removed and "--hostname my-rabbit" should be added to the extra parameters box. "name" (labelled as "Instance Name" in the template) can be removed. This is set by unraid using the "Name" entry. "rabbitmq" (labelled as "Image" in the template) can be removed. This is set by unraid using the "Repository" entry. Persistent data can be achieved by linking the container path: /var/lib/rabbitmq to an unraid appdata path such as: /mnt/user/appdata/rabbitmq/data/ Doing so will allow data to survive even a container delete and re-create. The linked directory can be empty on first start. It's also critical the hostname is set in extra parameters or else the data will keep getting reset. Persistent config was a bit more tricky. I linked the container path: /etc/rabbitmq to an unraid appdata path: /mnt/user/appdata/rabbitmq/config/ and then made the files, which can be done with something like this from within unraid: mkdir /mnt/user/appdata/rabbitmq/config/conf.d/ echo "[rabbitmq_management,rabbitmq_prometheus]." > /mnt/user/appdata/rabbitmq/config/enabled_plugins echo "loopback_users.guest = false" >> /mnt/user/appdata/rabbitmq/config/conf.d/10-defaults.conf echo "log.console = true" >> /mnt/user/appdata/rabbitmq/config/conf.d/10-defaults.conf You could also run the container first and copy out the /etc/rabbitmq path to get those files. The defaults file comes from: https://github.com/docker-library/rabbitmq/blob/master/10-defaults.conf, but for reference see: https://www.rabbitmq.com/configure.html#example-config Finally proper permissions can be set with: chown -R 999:999 /mnt/user/appdata/rabbitmq/data chown -R nobody:users /mnt/user/appdata/rabbitmq/config chmod -R 777 /mnt/user/appdata/rabbitmq/config
  4. Thank you to tmchow for bringing this docker image to the community apps. For many cases n8n needs to be accessible externally (for webhooks, oauth2, etc), the best way is to add the following docker vars: N8N_BASIC_AUTH_ACTIVE=true N8N_BASIC_AUTH_USER=user N8N_BASIC_AUTH_PASSWORD=pass