Extracting the Cups Drivers on Ubuntu 7.10

[HodgeHi]

If you know there's meant to be something in there, do the following:

 

shell: updatedb

shell: locate filename

 

and it should point to the correct location of the file.

 

I don't know what files should be there as i have never done this before. I assumed that the drivers folder would be there.

 

Can i point filename to the folder/directory name instead?

[ahuxham]

You sure can.

 

For example Samba, dependant of system, creates its winbindd pipe in different places, so I did a "locate d_pipe" and lo and behold it found the correct location of the folder.

 

Its just a text search, either folder locations, filenames, etc etc

[HodgeHi]

Thanks.

 

I just run the command like so:

 

mark@ubuntu:~$ locate /etc/samba/drivers
mark@ubuntu:~$ sudo ls /etc/samba
dhcp.conf  gdbcommands  smb.conf  smb.conf~  smbpasswd

 

It returned me to the command prompt. I listed the directory and there wasn't anything there. Am i using it incorrectly? I presume i am.

[ahuxham]

Thanks.

 

I just run the command like so:

 

mark@ubuntu:~$ locate /etc/samba/drivers
mark@ubuntu:~$ sudo ls /etc/samba
dhcp.conf  gdbcommands  smb.conf  smb.conf~  smbpasswd

 

It returned me to the command prompt. I listed the directory and there wasn't anything there. Am i using it incorrectly? I presume i am.

 

You'll want to run "locate drivers" thus showing all locations of that folder, it may have placed it in a different location, or surely, these "drivers" are the ones you've just downloaded, wouldn't you need to point that toward the location where they are shared?

[HodgeHi]

This is where i get lost. You see i have been following Ric_ guide on the installation. He suggests extracting the cups drivers into /usr/share/cups/drivers and then chown -R admin:admin /etc/samba/drivers.

 

If what you say is the case then i can copy or even move the file to the samba location.

 

I may just do that anyway.

 

PS thanks for all the help guys. Please bear with me, i think i'm nearly there.

[ahuxham]

This is where i get lost. You see i have been following Ric_ guide on the installation. He suggests extracting the cups drivers into /usr/share/cups/drivers and then chown -R admin:admin /etc/samba/drivers.

 

If what you say is the case then i can copy or even move the file to the samba location.

 

I may just do that anyway.

 

PS thanks for all the help guys. Please bear with me, i think i'm nearly there.

 

 

"mkdir /etc/samba/drivers" than copy the original drivers in there, chown the folder as you would, and leave both locations chowned to domain admins, that way either one will work, if you're not sure about location.

 

Another point may be to send him a PM and suggest the guides not very clear on those steps.

[HodgeHi]

I'm only trying to get a fully integrated system working between Windows and OS X.

 

I have set up a Print Queue on the Ad to print to the Linux print server and it prints fine with no authentication needed.

 

I have also set up a print queue on the OS X side and printed from there but it still asks for authentication. It should already have been pre-authenticated as it should be running kerberos authentication from the AD but seems to still require a user to type in their name and password in OS X.

 

Here is how i set up my print queue on os x:

[Geoff]

Rics instructions are wrong. You should:

 

chown -R administrator:"domain admins" /usr/share/cups/drivers

[HodgeHi]

OK. I will change the command.

 

Anyone else know why my OS X are requesting authentication when they really shouldn't need to?

 

 

Thanks.

[ahuxham]

OSX > Linux (Ubuntu) > Windows.

 

theres kerberos communication between ubuntu and windows, as you've clearly set this up, not sure how an un-authenticated osx request would be handled by the ubuntu machine, when its expecting a response from the PDC.

 

Windows user prints to Linux, linux machine asks windows if this is ok, response ok via kerberos/samba/winbind, document printers.

 

OSX user prints to Linux, linux machine asks .... if this is ok, response unknown, as the linux machine doesnt know what to-do with the request?

[Ric_]

There is a reason that I added the files there... isn't it something to do with the cupsaddsmb command? (or perhaps Ubuntu is slightly different to Debian in this respect?)... it does seem like a random place for me to put them

[Geoff]

Anyone else know why my OS X are requesting authentication when they really shouldn't need to?

 

Because your Linux machine is now part of your Windows domain. As such, it expects clients to be on the same domain and also requires clients to authenticate.

 

Are your Macs on the domain as well?

[HodgeHi]

Yes. They are bound to the AD as well. My AD users are logging in fine onto the client i have set up.

 

The OD server is also set up and configured like the AD-OD integration paper.

 

I have just found this though

 

Configuring CUPS to Use Kerberos

 

Once you have configured Kerberos on your system(s), you can then enable Kerberos authentication by selecting the Negotiate authentication type. The simplest way to do this is using the cupsctl(8) command:

 

cupsctl DefaultAuthType=Negotiate

You can also enable Kerberos from the web interface by checking the Use Kerberos Authentication box and clicking Change Settings:

 

http://localhost:631/admin

After you have enabled Kerberos authentication, add AuthType Default lines to the policies you want to protect with authentication, for example:

 

Listing 1: Remote Printer Operation Policy

 

1

2 # Job-related operations must be done by the owner or an

administrator...

3

Restart-Job Purge-Jobs Set-Job-Attributes

Create-Job-Subscription Renew-Subscription

Cancel-Subscription Get-Notifications Reprocess-Job

Cancel-Current-Job Suspend-Current-Job Resume-Job

CUPS-Move-Job>

4 AuthType Default

5 Require user @OWNER @SYSTEM

6 Order deny,allow

7

8

9 # Require authentication when creating jobs

10

11 AuthType Default

12 Require valid-user

13 Order deny,allow

14

15

16 # All administration operations require an administrator

to authenticate...

17

CUPS-Add-Class CUPS-Delete-Class CUPS-Set-Default>

18 AuthType Default

19 Require user @SYSTEM

20 Order deny,allow

21

22

23 # All printer operations require a printer operator

to authenticate...

24

Set-Printer-Attributes Enable-Printer Disable-Printer

Pause-Printer-After-Current-Job Hold-New-Jobs

Release-Held-New-Jobs Deactivate-Printer Activate-Printer

Restart-Printer Shutdown-Printer Startup-Printer

Promote-Job Schedule-Job-After CUPS-Accept-Jobs

CUPS-Reject-Jobs>

25 AuthType Default

26 Require user varies by OS

27 Order deny,allow

28

29

30 # Only the owner or an administrator can cancel or

authenticate a job...

31

32 Require user @OWNER @SYSTEM

33 Order deny,allow

34

35

36

37 Order deny,allow

38

39

 

What would i need to change to get this to work in my domain if my realm was CRONEHILLS.SANDWELL.SCH.UK Would it be similar to the Cups.conf file in the guide?

 

I am slowly looking to make things simpler again. If i can edit this config file to suit then it may just get things going for me the way i would like them to be.

 

Hope someone can save my day.

 

PS. We need a superhero emoticon

[HodgeHi]

My Setup is like this at the moment:

 

AD is configured as the PDC. This holds the kerberos realm.

 

OD is connected to the AD and is part of the kerberised realm hosted by the AD. It is used to host the home profiles and also manage the OS X users.

 

Linux is also connected to the AD and uses the AD Kerberos realm. This allows users from the AD to print to the queues without authentication.

 

The AD users print to the print queue that is connected to the Ubuntu Server

OD users print to the queue on the ubuntu server.

[HodgeHi]

OK, so i am at the point where i need to install and configure PyKota onto my Ubuntu Linux box. So where do i start?

 

I have looked on the PyKota website and it says i need to pay to get the package. But there is a download on the subversion repository.

 

What do i need to do to install the PyKota software. The steps in the install guide says to use dpkg -i but i don't know what the filename is?

 

I thought this was going to be a little easier than this but it is certainly a job for someone with a little more knowledge of how Linux works and not really for the adventurous types as they would perhaps give up and go home

[Ric_]

To get the packages you can get them out of SVN or stump up a donation (it's about £10 IIRC). It really is worth it's weight in gold so I donated and downloaded the packages.

 

Once you have the packages, it is VERY easy to install.

[HodgeHi]

Unfortunately i don't have a credit card to donate :-(

[Ric_]

I can send you the PyKota packages (the license alllows this) but I would urge you to get somebody to cough up some cash. You'll see how great the app is.

 

Drop me an email address and I'll see what I can do.

[HodgeHi]

I appreciate the help.

[Geoff]

Once you pay, you get access to the download section. Which includes .deb files. You can install the .deb files with 'dpkg -i'.

[HodgeHi]

Just to tidy up the thread i have since decided to move on from the Linux and PyKota installation due the amount of issues i was running into and the time scale i have. I decided to give Paper Cut a try and it does what i need without the issue of needing kerberised print queues. Instead of monitoring 1 set of queues on one server, Paper cut allows me to monitor 2 sets of queues on individual servers but hosting the quota system in one location.

 

Thanks for all the help though. It was most appreciated and if nothing else, it was a great Linux experience and certainly helped me understand it a bit better. I will most likely look at implementing something on a Linux server in the future, maybe a File server since it seemed pretty quick and easy to get this part up and running.

 

[ahuxham]

Just to tidy up the thread i have since decided to move on from the Linux and PyKota installation due the amount of issues i was running into and the time scale i have. I decided to give Paper Cut a try and it does what i need without the issue of needing kerberised print queues. Instead of monitoring 1 set of queues on one server, Paper cut allows me to monitor 2 sets of queues on individual servers but hosting the quota system in one location.

 

Thanks for all the help though. It was most appreciated and if nothing else, it was a great Linux experience and certainly helped me understand it a bit better. I will most likely look at implementing something on a Linux server in the future, maybe a File server since it seemed pretty quick and easy to get this part up and running.

 

 

It's always handy getting used to the system and how it works before starting a test, albeit step-by-step.

 

Nothings stopping you grabbing an old P3 system with minimal ram, taking it home and installing Ubuntu on it and messing around. It's how I learnt.. by butchering systems together =]