Here's my winders gripe for the day..

Story: Five Tired Old Myths About GNU/LinuxTotal Replies: 9
Author Content
techiem2

May 24, 2007
6:29 AM EDT
On the topic of how wonderful windows hardware support is:

We just setup a new XP Pro laptop at the college for someone who works from home. We set everything up, installed the same model printer, etc. Then we brought her in to setup her other stuff like her silly iPod and whatnot and showed her how to hook it all up. So she goes home and...her printer won't work. Windows is trying to reinstall it. So I tell her to go ahead and login as admin and let it. She logs in as admin, lets windows reinstall the printer, and it finally works. Of course windows had created a new printer instance which she had to use. Same model printer, same usb port, different physical printer. And just because of that windows has to reinstall the drivers and make a new printer instead of just using the already created instance of the printer. What is up with that?

Aladdin_Sane

May 24, 2007
7:45 AM EDT
Um, Would not KDE/GNOME do the same thing, given the same circumstance?

Not that I don't understand that any Windows gripe requires an outlet...
techiem2

May 24, 2007
8:38 AM EDT
I guess that would be the question.

What happens if you hook up a usb printer to linux, set it up in cups, then replace it with the identical model on the same port?

Aladdin_Sane

May 24, 2007
9:50 AM EDT
I missed the "Same model printer" in your original post.

I know that Win95 would reboot itself during first boot after keyboard change for no adequately explained reason.

Here are some possibilities:

1) WinXP is reading a serial number from the printer ROM.

2) WinXP is using its famous existential PnP to enumerate devices differently on each boot.

3) The 2 printers BIOS's are a different revision.

4) The printer is a "Software" or "Win" printer and has a buggy driver.

5) The printer is a "Software" or "Win" printer and has a non-buggy driver.

--

Famous Linux USB printer lore:

If Linux 2.6 USB driver gets an "unknown" USB printer error, it will return "printer on fire" to the OS. See

drivers/usb/class/usblp.c in the source.

With all these new Ubuntu users coming on line, there's a distress lawsuit just waiting to happen. I wonder if Dell knows...

Use

grep -r -i "on fire" *

on the source code to see what all can "burn" in Linux.
techiem2

May 24, 2007
12:13 PM EDT
FWIW, the printer is an HP LaserJet 1020.
frazw

May 25, 2007
7:48 AM EDT
Does anyone know how USB devices present themselves? I thought it might be due to the USB controller chip having a unique id. If it's a new id will it consider it as hardware which has not been previously installed even though the same model may have been connected before? I've seen windows regard identical models of the same memory stick as new hardware and set about installing them. I think the difference could be that the linux kernel has support enabled and sees all e.g. usb memory sticks as a usb memory stick rather than windows thinking it is usb memory stick 0001. I dunno just a thought
Aladdin_Sane

May 25, 2007
7:59 AM EDT
>>Does anyone know how USB devices present themselves? I thought it might be due to the USB controller chip having a unique id. If it's a new id will it consider it as hardware which has not been previously installed even though the same model may have been connected before?

Yes, this more-or-less what I meant by a serial number in the printer. No, the controller chip is on the system, not the printer.

>>I've seen windows regard identical models of the same memory stick as new hardware and set about installing them. I think the difference could be that the linux kernel has support enabled and sees all e.g. usb memory sticks as a usb memory stick rather than windows thinking it is usb memory stick 0001.

Hmm, this is similar to the enumeration issue I spoke of. But not quite. Sys device (like USB controller) enumeration occurs on boot. USB device enumeration is hot-plug, and can occur at any time, I believe.

In Debian, you can keep track with lspci and lsusb to see what happens.

In Windows, you cannot.
nevva

May 25, 2007
8:46 AM EDT
Could it be that she plugged it in on a different USB port? That would be my guess..
techiem2

May 25, 2007
9:00 AM EDT
Nope. We made sure she used the same port. *shrug* I could understand setting it up again if you use a different port. I'm guessing it was some sort of device ID/serial number type issue as has been mentioned.

amby

May 25, 2007
2:09 PM EDT
frazw: "Does anyone know how USB devices present themselves?"

[17372955.748000] usb 4-1: new full speed USB device using uhci_hcd and address 3 [17372955.936000] usb 4-1: configuration #1 chosen from 1 choice [17372955.940000] drivers/usb/class/usblp.c: usblp0: USB Bidirectional printer dev 3 if 0 alt 0 proto 2 vid 0x04E8 pid 0x323A

Posting in this forum is limited to members of the group: [ForumMods, SITEADMINS, MEMBERS.]

Becoming a member of LXer is easy and free. Join Us!