[cups] Brother vs cups, 0,1 in favor of cups

Gene Heskett gheskett at shentel.net
Sat Nov 3 19:14:07 PDT 2018 

On Saturday 03 November 2018 19:26:12 Brian Potkin wrote:

> On Sat, 3 Nov 2018 at 22:18, Gene Heskett <gheskett at shentel.net> wrote:
> > On Saturday 03 November 2018 16:04:08 Brian Potkin wrote:
> > > On Sat, 3 Nov 2018 at 17:21, Gene Heskett <gheskett at shentel.net> 
wrote:
> > > > To continue this thread, I've gone to the brother site and found
> > > > an installer script that pulled all the latest driver debs &
> > > > what not, over-writing the original i386 installs with amd64
> > > > versions. That of
> > >
> > > That script is amazing, considering that Brother say
> > >
> > >  > I'm using a Linux 64 bit edition. Can I use the Brother Linux
> > >
> > > printer drivers?
> >
> > I don't see why not. Shouldn't be a problem as long as its plugged
> > into a usb port. Where its getting covered with nasty hair is when
> > you are using cat5 and an ipv4 address. I rebooted just to get the
> > startup messages, and some in the (I presume) udev camp has decided
> > that only one device per address is to be allowed. Scattered thru
> > the syslog:
> >
> > Nov  3 13:58:50 coyote avahi-daemon[4577]: Host name conflict for
> > "scanner.local", not established. Nov  3 13:58:50 coyote
> > saned[4684]: saned (AF-indep+IPv6) from sane-backends 1.0.22
> > starting up Nov  3 13:58:50 coyote saned[4684]: do_bindings: [0]
> > bind failed: Address already in use
> >
> > later in the log:
> >
> > Nov  3 13:58:53 coyote xinetd[4771]: Reading included configuration
> > file: /etc/xinetd.d/amanda [file=/etc/xinetd.conf] [line=14] Nov  3
> > 13:58:54 coyote xinetd[4771]: failed to parse raspberrypi.coyote.den
> > [file=/etc/xinetd.d/amanda] [line=8]
> >
> > but I've already fixed that, the pi's hostname got changed, then the
> > door- bell woke me up, a dinner delivery from our church.  Wayyyyy
> > to sweet for us diabetics so some of it will go in the square can.
> >
> > To get back on track:
> >
> > Nov  3 13:58:54 coyote xinetd[4771]: bad service attribute: ipv6
> > [file=/etc/xinetd.d/saned] [line=8] Commented that line out.
> >
> > Then 25 lines later:
> > Nov  3 13:58:54 coyote xinetd[4771]: bind failed (Address already in
> > use (errno = 98)). service = sane-port Nov  3 13:58:54 coyote
> > xinetd[4771]: Service sane-port failed to start and is deactivated.
> > Nov  3 13:58:54 coyote xinetd[4771]: bind failed (Address already in
> > use (errno = 98)). service = ssh Nov  3 13:58:54 coyote
> > xinetd[4771]: Service ssh failed to start and is deactivated.
> >
> > But ssh is working just fime, I am logged into 6 other local
> > machines, and have 3 of them also sshfs mounted that mount with
> > keys.
> >
> > I might be able to find it with a usb cable but I'd have to double
> > the length of the cable as the first 40"+ of it is inside this
> > printer. So it would take well over the maximum unbuffered usb cable
> > length of 5 feet to reach it from any of the usb hubs laying around.
> >  The only one with matching connectors is holding a 7 port hub up in
> > the air by about a foot. Roughly the same length as the height of a
> > mahogany paper and ink storage (and anything else I want out of
> > sight) cabinet I made, and that the MFC is sitting on the top of.
> >
> > And it looks like it might be working, from dmesg when I plugged it
> > in: [14060.260317] usb 1-3.4.3: new high-speed USB device number 21
> > using ehci-pci [14060.374449] usb 1-3.4.3: New USB device found,
> > idVendor=04f9, idProduct=02f5 [14060.374453] usb 1-3.4.3: New USB
> > device strings: Mfr=1, Product=2, SerialNumber=3 [14060.374456] usb
> > 1-3.4.3: Product: MFC-J6920DW
> > [14060.374458] usb 1-3.4.3: Manufacturer: Brother
> > [14060.374459] usb 1-3.4.3: SerialNumber: BROG5F229909
> > [14060.375378] usblp 1-3.4.3:1.0: usblp3: USB Bidirectional printer
> > dev 21 if 0 alt 0 proto 2 vid 0x04F9 pid 0x02F5 [14060.375618]
> > usb-storage 1-3.4.3:1.3: USB Mass Storage device detected
> > [14060.375702] scsi16 : usb-storage 1-3.4.3:1.3
> > [14061.392796] scsi 16:0:0:0: Direct-Access     Brother  MFC-J6920DW
> >      1.00 PQ: 0 ANSI: 2 [14061.393089] sd 16:0:0:0: Attached scsi
> > generic sg6 type 0 [14061.472613] sd 16:0:0:0: [sdf] Attached SCSI
> > removable disk
>
> If there is any relevance here to the existence of 64-bit Brother
> printer drivers, it
> is over my head.
>
> > Yup:
> >
> > gene at coyote:~/Downloads/SDC-upgrade/CoCo$ scanimage -L
> > device `brother4:bus2;dev1' is a Brother MFC-J6920DW USB scanner
> >
> > And xsane connects, ready to scan, so that is butt ugly but it
> > works. But anyone disturbing this midden heap will destroy it all.
> > One of that guy Murphy's laws.  Sigh.
>
> The scanner is discovered. Who could want for more?

For it to be discovered where it also lives on the end of a 10 foot piece 
of cat5, at the same address as the printer lives at, and where it has 
Just Worked(TM) for 2+ years.

Some know it all, presumably in the udev camp has decreed that two 
functions, a printer and a scanner on 2 separate port numbers of a tcp 
path to the same address, should not be allowed. One of the last things 
I got from the wheezy repos just a  week, maybe 2 back was a udev update 
thru the security channel. I expect thats when the scanner actually 
died. I was not aware of it as I don't use it that often.
>
> > > > Yes. Brother printer drivers are created and optimized for 32
> > > > bit version of Linux, but those can be used for 64 bit Linux
> > > > also. Some additional steps are required.
> > >
> > > They then go on to give instructions.
> >
> > Yeah, what they DO give comes with bloodstains from their legal
> > people, but if they say it, its quite likely so.
> >
> > > There are no 64-bit drivers from Brother but you seem to have
> > > hunted them out. No wonder ordinary users find printing difficult.
> > > You really must reveal your secret
> > > technique!
> > >
> > > [Snip]
> >
> > No secret, just point FF at brothers site, and drill down to your
> > printer model, and in my case theres at least two more subdirs, but
> > you'll eventually get a list of debs for both cpu families, click on
> > the i386 versions and follow the links to save them locally.
> >
> > I made a subdir in ~/Downloads/brother-drivers:
> > -rw-r--r-- 1 gene gene     1558 Nov  2 20:55
> > brother-udev-rule-type1-1.0.2-0.all.deb -rw-r--r-- 1 root root   
> > 87658 Sep 12 02:33 brscan4-0.4.6-1.amd64.deb -rw-r--r-- 1 gene gene 
> >   75124 Nov  2 20:19 brscan4-0.4.6-1.i386.deb -rw-r--r-- 1 root root
> >    50852 Sep  5  2013 brscan-skey-0.2.4-1.amd64.deb -rw-r--r-- 1
> > gene gene    44942 Nov  2 20:54 brscan-skey-0.2.4-1.i386.deb
> > -rw-r--r-- 1 gene gene 24902333 Nov  2 22:37
> > mfc6920dw_use_busr_lel566001_a.pdf -rw-r--r-- 1 root root    15348
> > Nov  2 20:37 mfcj6920dwcupswrapper-3.0.0-1a.i386.deb -rw-r--r-- 1
> > root root    15398 Jan 10  2014
> > mfcj6920dwcupswrapper-3.0.0-1.i386.deb -rw-r--r-- 1 gene gene   
> > 22764 Nov  2 20:21 mfcj6920dw_cupswrapper_GPL_source_3.0.0-1.tar.gz
> > -rw-r--r-- 1 root root  2957378 Nov  2 20:37
> > mfcj6920dwlpr-3.0.0-1a.i386.deb -rw-r--r-- 1 gene gene  2957186 Nov 
> > 2 21:02 mfcj6920dwlpr-3.0.0-1.i386.deb
>
> There are no 64-bit printer drivers.

Not that are easy to spot in the listing above, thoroughly trashed by 
wordwrapping an inserted > > but if you care to look at that list again, 
you will find amd64 scanner drivers. My installing the i386 versions 
overwrote the amd64 versions in place. And I've deleted the amd64 
packages for the printer, but if you go look at
<https://support.brother.com/g/s/id/linux/en/evaluation.html?c=us&lang=en&prod=mfcj6920dw_us_eu_as&redirect=on>
you will find some amd64 packages, which I assume are what the install
script used when it assumed, after finding I had a quad core phenom,
was a 64 bit install. Its mistake of course was in not checking to see
which version of libc6 was installed.
>
> > So I didn't have to wade thru all the trash to find them. You just
> > have to keep your shovel handle clean and keep digging.
> >
> > > > But scanimage -l still can't find the scanner even with the i386
> > > > versions of its drivers re-installed.  Its address on the local
> > > > net pings just fine, its the same as the printer, but sending FF
> > > > to that address gets no mention of there also being a scanner at
> > > > that same address.
> > >
> > > "-l" is a typo, we suppose. Brother do provide instructions for a
> > > scanner setup.
> >
> > -l and -L are the same except -L digs a deeper trench as root.
>
> Not on anyone else's system.
>
read the man page.

> > Anyway, my problems seem to have been reduced to a not quite perfect
> > but usable state, so until the next bug comes crawling out of the
> > plastic with a target painted on its back, I can get on to other
> > things.
>
> There are no bugs. Just your own ineptitude.

My ineptitude?

>
> > Thanks everybody.
> >
> > And maybe, just maybe, I've helped someone else battling with
> > generally good, brother products.  That would be nice, I like
> > flowers but dasn't dare breath on them. :(
>
> The product is not the problem.

The install script assumed the wrong linux install.

That is NOT my ineptitude, but Brothers. And I am not willing to shoulder 
any blame for that from somebody who probably made their first linux 
install years after mine, of red hat 5.0 in 1998.  You, Brian Potkin, 
weren't looking to help, only to criticize. If you can't help, then the 
list is better off if you STFU. I asked the question, then, 
understanding that the answer because this MFC was an oddball able to 
scan and  print tabloid sized paper, was not going to be served up by 
some one who had been there and done that, I seem to have a talent for 
going places and getting things done that is, shall we say eclectic, 
exploring some of the darker corners of things, so I got busy and found 
my own answers and posted the results to the list so the next person 
might be helped. Did you do that Brian Potkin, or are you just blowing 
your own horn?


-- 
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>

More information about the cups mailing list