Bug 186880

Fixing version.

What CPU type are you using?  Is it x86_64 perhaps?

AMD X2

Should be fixed in the foomatic I built on Friday, 3.0.2-34.

Created attachment 126832 [details]
part of /var/logcups/error_log

Does this work now?

It does not work yet. 
I can add few things.
I installed on my (centrino) Laptop FC5 and I plug my printer on the usb device.
I can print from my laptop. Moreover,  from my Opteron computer, I see the my
printer with cups and I can print by using the network and hence the printer
connected to my laptop.
This means both that cups works on X86_64 for printers on the network
but not connected locally 
I also notice a curious message in /var/log/messages when I plug my printerApr 
3 17:25:41 anacan4 kernel: usb 1-1: new full speed USB device using ohci_hc    
                                       d and address 3
Apr  3 17:25:41 anacan4 kernel: usb 1-1: configuration #1 chosen from 1 choice
Apr  3 17:25:41 anacan4 kernel: drivers/usb/class/usblp.c: usblp0: USB Bidirecti
                                           onal printer dev 3 if 1 alt 0 proto 2
vid 0x04B8 pid 0x0802
Apr  3 17:25:41 anacan4 kernel: usbcore: registered new driver usblp
Apr  3 17:25:41 anacan4 kernel: drivers/usb/class/usblp.c: v0.13: USB Printer De
                                           vice Class driver
Apr  3 17:25:50 anacan4 hpiod: ParDevice::nibble_read failed: Input/output error
                                           
I scanned Google and I found that such a message occurs in bugzilla, #183433. 

the typical error_log message is
E [03/Apr/2006:17:30:16 +0200] [Job 10] Loading NimbusRomNo9L-Regu font from
/usr/share/fonts/default/Type1/n021003l.pfb... 3244808 1931530 1780296 478707 1
done.
E [03/Apr/2006:17:30:16 +0200] [Job 10] Using NimbusRomanNo9L-Regu font for
NimbusRomNo9L-Regu.
E [03/Apr/2006:17:30:16 +0200] [Job 10] Loading NimbusSanL-ReguCondItal font
from /usr/share/fonts/default/Type1/n019063l.pfb... 3399192 2079248 1800480
502607 1 done.
E [03/Apr/2006:17:30:18 +0200] [Job 10] KID3 exited with status 0
E [03/Apr/2006:17:30:18 +0200] [Job 10] KID4 exited with status 9
E [03/Apr/2006:17:30:18 +0200] [Job 10] Process dying with "error closing
*main::STDOUT", exit stat: 9
E [03/Apr/2006:17:30:18 +0200] [Job 10] error: Relais bris�(pipe) (32)
E [03/Apr/2006:17:30:18 +0200] [Job 10] error closing *main::STDOUT
E [03/Apr/2006:17:30:18 +0200] [Job 10] Renderer exit stat: 9
E [03/Apr/2006:17:30:18 +0200] [Job 10] KID3 finished
E [03/Apr/2006:17:30:18 +0200] [Job 10] Renderer process finished
E [03/Apr/2006:17:30:18 +0200] [Job 10] Killing process 3419 (KID3)
E [03/Apr/2006:17:30:18 +0200] [Job 10] Process dying with "Error closing
renderer", exit stat: 9
E [03/Apr/2006:17:30:18 +0200] [Job 10] error: Mauvais descripteur de fichier (9)
E [03/Apr/2006:17:30:18 +0200] [Job 10] Error closing renderer
E [03/Apr/2006:17:30:18 +0200] PID 3416 (/usr/lib/cups/filter/foomatic-rip)
stopped with status 9!

Thanks for your interest
Sincerely yours, Pascal 

Should be fixed in cups-1.2-0.2.rc1.6 (devel) and cups-1.2-0.2.rc1.2.6 (FC-5
updates-testing).

The x86_64 devel packages are here:

  ftp://people.redhat.com/twaugh/tmp/cups-devel/x86_64/

if you would like to try them before tomorrow's rawhide tree is pushed.

Created attachment 127303 [details]
/var/log/cups/error_log

The new version of cups does not fix the problem, but (see log file) I noted that
the CUPS environnement variables do not start with /usr/lib64 but with /usr/lib
and could explain my problem. Since the Espon CX3200 is also an scanner, I tried
it and it works.
If the problem is due to the environnement variables, what the configuration file
where I can change the path
Thanks again for your help

The paths '/usr/lib/cups/backend' and '/usr/lib/cups/filter' are correct in CUPS
1.2, even on x86_64 -- those directories always contain executables and not
libraries.

The problem is this line in the log:

E [04/Apr/2006:18:02:26 +0200] [Job 31] Unable to open USB device
"usb:/dev/usb/lp0": Permission denied

Please show me the output of this command:

ls -l /dev/usb/lp0

ls -l /dev/usb/lp0
crw------- 1 viot root 180, 0 avr  4 13:34 /dev/usb/lp0

I do not undertand why I am the owner???

Please attach /etc/security/console.perms.  You may need to do this as root first:

cat /etc/security/console.perms >/tmp/console.perms
chmod a+r /tmp/console.perms

It certainly isn't normal that you are the owner.  The ownership should be
root:lp, and this is the reason you cannot print.

Created attachment 127306 [details]
console.perms file

After some minutes, I have
ls -l /dev/usb/lp0
crw------- 1 viot lp 180, 0 avr  4 18:54 /dev/usb/lp0

but 
lpq
lw is not ready
Rank	Owner	Job	File(s) 			Total Size
1st	viot	35	hosts				1024 bytes

???

Well, the lp group still can't read or write to that node.

What do you mean 'after some minutes' -- did it just change on its own?

On the FC-5 machine I have here, the permissions are correct, so I think this
must be some local modification you have made.

I did not change the permissions, but I only removed the job of testpage and
sent a job from my account.
Anyway, I just found a workaround and my printer works finally
chmod o+rw /dev/usb/lp0
chmod g+rw /dev/usb/lp0

The error_log gives then
 [04/Apr/2006:22:54:55 +0200] [Job 38] KID4 exited with status 0
E [04/Apr/2006:22:54:55 +0200] [Job 38] Renderer exit stat: 0
E [04/Apr/2006:22:54:55 +0200] [Job 38] Renderer process finished
E [04/Apr/2006:22:54:55 +0200] [Job 38]
E [04/Apr/2006:22:54:55 +0200] [Job 38] Closing foomatic-rip.
D [04/Apr/2006:22:54:55 +0200] PID 7334 (/usr/lib/cups/filter/foomatic-rip)
exited with no errors.
D [04/Apr/2006:22:54:57 +0200] PID 7335 (/usr/lib/cups/backend/usb) exited with
no errors.

I do not know if theses permissions remain correct from a point of view of  a
security level, but it works.
Thanks again for your help!
I am glad of Fedora (our Beowulf (Bi-opteron) is using Fedora x86-64
http://www.lptl.jussieu.fr/) and our Nas server Redhat Enterprise.
Thanks again