You would expose the port to your host which makes the db acessible by anything running on the host, docker or native. Something like

`port

• 5432:5432 `

But I would recommend running a dedicated db for each service. At least that’s what I do.

• Simpler setup and therefore less error-prone
• More secure because the db’s don’t need to be exposed
• Easier to manage because I can independently upgrade, backup, move

Isn’t the point about containers that you keep things which depend on each other together, eliminating dependencies? A single db would be a unecessary dependency in my view. What if one service requires a new version of MySQL, and another one does not yet support the new version?

I also run all my databases via a bind mount

`volume

• ./data:/etc/postgres/data…`

and each service in it’s own directory. E.g. /opt/docker/nextcloud

That way I have everything which makes up a service contained in one folder. Easy to backup/restore, easy to move, and not the least, clean.

For me it’s Borg backup for Nextcloud an all the other servers

For a while it’s just data in, which it handles really well. But it really started to shine for me when I needed to find some of the documents. OCR and their search works very well for me.

There are also some interesting thoughts in here: https://skerritt.blog/how-i-store-physical-documents/

What’s the error?

Can you run docker-compose logs

Here’s mine: https://cloud.zell-mbc.com/s/Ac5KQTTxcWNYbNs

I tried to add file it to this post but formatting got completely messed up, hence a link.

Before you run docker-compose you need to change the paperless-app volumes to fit your requirements and set up the variables in .env

Let me throw in Paperless NGX, https://github.com/paperless-ngx/paperless-ngx

:-)

But seriously, I was wondering about the requirement to shutdown the VM’s and couldn’t come up with a solid reason? I mean, even if QEMU/KVM/Kernel get replaced during a version upgrade or a more common update, all of these kick in only after the reboot? And how’s me shutting down VMs manually different from the OS shutting down during a reboot?

I know I am speculating and may not have the fill picture, probably a question for the Proxmox team, there may be some corner case where this is indeed important.

By the way, Mexican or US black strat? :-)

Like you I have OPNsense in a VM on one of my PVEs. But I only made sure the nigthly VM back up ran and didnt even bother shutting down the VMs during the upgrade. The VMs got restarted during the final reboot, as the would with every other reboot, and I was back in business.

<<so I don’t want to install the nginx container in the sample compose either>> Thought the same initially but it turned out both nginx instances are required. I have a working setup with an installed nginx instance + certbot and an almost unmodified docker-compose.yml The modifications I made were changing container names (so they fit my standard) and the http port

The lemmy documentation https://join-lemmy.org/docs/administration/install_docker.html recommends the following location block for the installed nginx:

  location / {
       proxy_pass http://localhost:LEMMY_PORT;
       proxy_set_header Host $host;
       include proxy_params;
}

This did not work for me! I had to comment the proxy_set_header line to make things work. With it I run into an error 400.

Here’s my little list:

Proxmox VE 1

1. OpnSense (VM)
   • Wireguard
   • Unbound
2. PiHole (LXE)

Proxmox VE 2

1. Dockerhost (LXE)

• Nextcloud
• Akkoma
• Lemmy
• Synapse
• Keycloak
• Dashy
• Paperless NGX
• Whoogle
• Grafana
• CCU Historian
• DMARC / TLS Reports
• Prometheus

RaspberryMatic Octoprint

eMail

• Postfix, Dovecot, etc.
• Ciphermail email encryption

I see that @stulli recomends the gutmensch docker-compose repo. That repo is using the techsneeze repos I mentioned above, in fact I use the gutmensch docker setup as well. Maybe you are also interested in a pull request in the techsneeze repo which adds support for TLS reports. It’s sitting there for months but for some reason doesn’t get merged. But it works just fine for me and others which commented on the pull request. I ended up forking the techsneeze repos and applying the merge request on my fork…

By the way, I wasn’t able to respond to @stullis cooment because somehow his comment didn’t get federated to my Lemmy instance, teething problems I guess :-)

I run this one for a few months now: https://github.com/techsneeze/dmarcts-report-parser

I arrived at the same conclusion. Been trying to connect from my Akkoma account but like you, figured out this doesn’t make a lot of sense. So here I am, posting from my 2 hour old Lemmy server :-)