openpgp4fpr:74EDD2488126072A9E9FD0C7348F97E620E0BA7A

  • 0 Posts
  • 13 Comments
Joined 9 months ago
cake
Cake day: February 11th, 2024

help-circle









  • Set up Paperless-ng on your server, generally with Docker, and map the Consume folder to wherever you want. Expose that on the network as a Samba or FTP share depending on your printer.

    Printers with a bit more than basic features allow you to “scan to target” and it’s basically designed to set up a Public share folder on windows and scan and your document just shows up on the computer. Same deal but map it to the consume folder on the server. Paperless automatically picks up and intakes anything dropped in the consume folder.

    So you end up just hitting Scan on the printer, the printer will dump the output into consume share via either samba or ftp, and Paperless automatically picks it up and puts it in the Inbox for ya.


  • tekeous@usenet.loltoLinux@lemmy.mlDocument Management System for Linux?
    link
    fedilink
    English
    arrow-up
    10
    arrow-down
    1
    ·
    11 months ago

    I use Paperless-ng and it’s great. Headlining feature is that it stores your documents in PDF in a plain folder which makes backing up easy. Another software that puts your documents in a database is no good unless it has its own backup method.

    Plus being on a network server means I can set up my printer to scan to there as a target, my phone to scan to there, computer, I can drop emails in the consume folder, etc. Easy peasy to get stuff in there.



  • I run Fedora Server on a blade server in a colo.

    Pros:

    • Cockpit is the GOAT 🐐
    • Descended from RHEL so everything is supported
    • Podman is the GOAT 🐐

    Cons:

    • Podman is getting worse, for instance they recently deprecated systemd generate and tell you to use Quadlet, for running pods, you need to use Kubernetes. This greatly complicates my workflow.
    • SELinux, while secure, and easy to troubleshoot with Cockpit, is a major pain in the ass that prevents most containers from accessing their data directories. It can be corrected but is extremely frustrating.
    • Quadlet is extremely inconsistent, I can copy the working unit file for a container and it works, change the name and variables for another container, and one launches but the other won’t start. One will have the wrong name. Stupid things, like putting the name in quotes, reloading, removing the quotes fixes it. I have harsh words for the idiot who deprecated systemd generate.
    • something like Tiddlywiki, their documentation will put you in /var/www but Fedora uses /usr/www or something. You get used to the Fedora things but you can end up on a goose chase sometimes.

    Those cons are starting to hit hard, and when I reimage this server next I’m probably going to Proxmox or Debian. Server 37 was good but I probably won’t bother with 39.