I got a new printer. Auto-discovered, added, and prints fine from Windows in 2 minutes. Auto discovered, added, and prints fine from OSX in 30 seconds. Auto-discovered and added on Linux, but trying to print results in "printer is unreachable at this time" - even after 50 re-installs, different configs, different drivers, different protocols.
I recognized that some computers were on different subnets, but couldn't figure out a pattern. It turns out that the printer has a setting called "Restricted Server List" and the default setting is null. Here's its description in the admin interface: "Comma-delimited list of IP addresses that are allowed to make TCP connections. Example: 157.184.0.0/24. where 0 is a wildcard and /24 is the network prefix."
It also has a setting called "Restricted Server List Options", set to block all ports by default. Here's its description: "By default, addresses not in the restricted server list will have all access blocked. When Block Printing Only is selected, addresses not in the restricted sever list will be blocked from printing only. When Block Printing and HTTP Only is selected, addresses not in the restricted server list will be blocked from printing and HTTP. "
Admin interface doesn't say this anywhere, but the default setting of no restricted servers apparently allows access from other networks, but not from the same network as the printer. I set the restricted servers to "192.168.132.0/24" and then I could access the printer admin web page and print to the printer from my Linux box, but not from any of the computers that were working before. So I set it to "192.168.0.0/16" and every computer on all subnets in my house can print and access the printer admin.
The default setting of no restricted servers was extremely non-intuitive in that it actually only restricted servers on the same subnet. And there was no such documentation.
What a crappy waste of 7 frickin' hours!
i say the designers + programmers made it so and you say users made it so. ... it seems we disagree here.
No. That's not what I said. I said the manufacturers not testing their equipment on Linux made it so, and more users would change that. Actually, looking at it again that isn't even true. This example has nothing to do with the operating system at all. It's caused by connecting with a computer on a different subnet (or I guess more accurately the same subnet as the printer), which would have happened even if the OS were Windows.