Debian 12. HP Laserjet Professional P1606dn
If it prints at all, it prints the top inch of the test page or just random binary. I have tried the recommended driver, the driverless driver, the Generic PCL 4/5 driver, the Generic PCL 6 driver. And probably others I am not remembering.
I am trying to print over Ethernet, but I am about to drag the printer over near my desk and print via USB.
Fortunately, I don’t have actual critical printing to do right now and I am only setting up a printer after installing Debian 12. BTW this means it is a fresh install of Debian 12 too.
I have been helpdesk support at a data center. I would not consider myself a dummy, but this is getting ridiculous. A task that should have taken all of 10 minutes has taken over 2 hours so far.
How are we ever going to get “The Year of Linux on the Desktop” if simple printing is and continues to be such a pain?
In my experience printing has always been pain on all operating systems, especially Windows.
The fact that you literally need to open up about half of all possible TCP and UDP ports in a firewall just to get all printing protocols over the network to work at all and that the vendors try to prevent you from using third party ink, and other consumables is good evidence that it is more the printer vendors fault than anyone else’s.
Perhaps we need an “open source” laserprinter?
I wonder if even just an open source firmware for printers could work.
Actually, I think it might take a whole “open source” company - the printer itself is sold assembled at a profit (Much more expensive than printers today) but it can be assembled from available plans and off-the-shelf parts. The control inside the computer would probably be something a lot like a Raspberry Pi.
The ink is provided at the actual cost, and the formula for the ink is available.
The firmware is downloaded from the host computer on printer power-up, so that can be fully open source- allowing the user to correct or add functionality. Perhaps driven by a high-level “printer language” that would make writing printer firmware easy to understand and update.
I just did a test with my Brother printer on Windows. I saw one port opened for the print protocol and four or five for various name service protocols (because I have a homelab and have screwed around with DNS a little too much, apparently). If you’re opening half of all possible TCP and UDP ports to get the printing protocol to work, you’re doing something wrong.
I said “for all print protocols”, as in all the ones network printers have to support to get all possible clients to work.
But why would you need to get all possible clients to work? Just get the ones that are actually on your network working. And don’t open your Internet-facing firewall unless for some bizarre reason you have to print from over the Internet (can’t really see a critical use case for this except for outliers).
Unless you’re running a web cafe or something and have to support random laptops that people bring in. At that point security is out the window anyway because who knows what will be going on your network.
If you have to print from a location away from your network ( like on a business trip ) us tailscale!
The point was that the protocols are badly designed and I was talking about firewalls between subnets.