Basic Printing on OpenBSD | Hacker News Reader

Basic Printing on OpenBSD

paedubucher.ch

58 pointspaedubucher27 comments

One small item that may make setup a little easier:

    # echo 'lpd_flags=""' >> /etc/rc.conf.local
    # /etc/rc.d/lpd restart

Can be replaced with rcctl(8):

    # rcctl enable lpd
    # rcctl restart lpd

Thanks for pointing this out! rcctl was added for OpenBSD 5.7, the manpage says. Absolute OpenBSD 2nd edition is older than that.

I'll update accordingly!

I remember lpd with some horror and was very glad when CUPS came along, made printing on Linux even easier than on Windows.

I recently also suffered through some horrors with Cups. A new version was released, and old drivers didn't work together with it. And the manufacturer (Samsung, which sold its printer business to HP) didn't provide updated drivers, of course. So I'm happy to have some reliable facility to at least print PostScript. But Cups is fine if you get a proper and current driver for your printer.

I imagine being a CUPS maintainer is not a good experience. They did something great, but probably only hear from end users when they are frustrated and not inclined to be civil.

Doesn’t Apple maintain CUPS?

Maybe? Wikipedia says

"In March 2002, Apple Inc. adopted CUPS as the printing system for Mac OS X 10.2.[6] In February 2007, Apple Inc. hired chief developer Michael Sweet and purchased the CUPS source code.[7] On December 20, 2019 Michael Sweet announced on his blog that he had left Apple.[8]"

I hope they recognize it’s a little bit important. A lot of their users still produce paper.

On his blog post he mentioned that there are still 2 other apple engineers working on CUPS, and that Apple still very much owns the code.

I can’t imagine it going away anytime soon.

They use it in every device they sell that can print (iOS and macOS), so it's pretty important for them.

That's not just cups but for every other printer/environment when a product loses support.

Printing in Windows is frustrating because the printer manufacturers insist on including their shite user space tools. It's often difficult to get "just the driver, thank you" without all the cruft.

The Windows driver setup seems fairly rich and well done otherwise. Though I do have to occasionally run services.msc to bounce the spooler when it has a stale idea of printer state...like "out of paper".

Often you can unzip them and look for a subfolder containing the .inf files and install those directly.

This was the best trick I had when I was using Windows, for ALL drivers I can extract. No bloatware, no weird apps on the background.

> Though I do have to occasionally run services.msc to bounce the spooler when it has a stale idea of printer state...

Especially for networked printers that is annoying! If I forgot to turn on my printer it always needs this trickery to get Windows to realise I've now actually turned it on.

Seems trivial, right? Like checking in every 30 seconds or so. I imagine there's a guy at Microsoft that keeps trying to check in the simple fix but keeps being thwarted.

Probably blocked by the guys that also said "Hey, you know what printers need? TONS of xml and SOAP. Yea. Printers should only be found by UPNP and controlled via SOAP."

Then WSD Printers were born.

I’m totally amazed: at my Dpt. people (windows) have to print using a specific driver because otherwise they “cannot”. Their prints get logged (uid).

I just lpr and that is it. Log: “user”.... It is a postscript printer.

Sometimes they cannot print because the authentication server is down...

I feel like lpd is quite a bit simpler than cups and I still use it.

If I had to use CUPS to print from a printer, I would get rid of the printer.

Maybe it's better now, but CUPS used to be a nightmare. I still avoid it whenever possible. Eric Raymond did a good job of describing just how bad it was.

http://www.catb.org/~esr/writings/cups-horror.html

Brings back memories of trying to get HPGL, PCL, or other non-postscript printers working with various flavors of Unix, lpd filter scripts, trying to get a form feed to spit out the paper, etc. Pretty sure there's an ugly Perl script of mine still doing this at a previous employer.

My friends and I used to create a little mayhem sending bursts of form feeds to the DEC band printer in the lab when the teacher was near it. It would shoot out paper 5 feet or so.

Reminds me a bit of the prank they played at AT&T by sending reverse form feeds to the printer, so that the victim had to insert the paper afresh.

> reverse form feeds

A form feed is 0x0C or ^L in ASCII. What the heck is a reverse form feed?

E2 80 8F 0C

By the way, I read it here: https://www.cs.princeton.edu/~bwk/memoir.html Possibly it isn't written "reverse form feed", but something the like.

It brought me memories of being 15 and hooking up a parallel (lpt) epson printer (a miniplotter really) to a netbsd machine, failing miserably and failing miserably.

Text printed fine (mostly), anything else resulted in pages and pages of ascii characters (lol).

Reply on news.ycombinator.com