Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Bug#995322: ipp-usb: prevents hp-toolbox and libsane-hpaio from working

25 views
Skip to first unread message

Giacomo Mulas

unread,
Sep 29, 2021, 12:40:04 PM9/29/21
to
Package: ipp-usb
Version: 0.9.17-3+b4
Severity: normal

Dear Maintainer,

Upon upgrading a computer to bullseye, I discovered that the
multifunction scanner/printer/fax HP Officejet Pro 8620 became
unusable as a scanner, and could not any more be accessed and managed
via hp-toolbox. The system did detect the printer to be connected
via USB, but was then unable to connect to it, be it via xsane (or
any other scanning software) or hp-toolbox. This very effectively
made the scanner unusable from the computer.
Upon perusing strace and going through the system logs (I suspected some
apparmor problem, I had to fix some after the upgrade), I found out instead
that the device was already in use by ipp-usb, which prevented xsane and/or
hp-toolbox to use it. Upon stopping ipp-usb, I was able to correctly
use the multifunction flawlessly.

Please, either fix this incompatibility or document it properly, so that
users of such USB-connected devices do not get them partially blocked
without any clue as to why. I actually did not explicitly select ipp-usb for
installation, it was carried over by the overall upgrade from buster
to bullseye (don't know exactly by what dependency/recommends).
Getting a message upon installation that this might cause problems
with xsane would have spared me a few hours of troubleshooting at the
moment I really needed the scanner to work (as it always did before).

Thanks in advance, best regards
Giacomo Mulas

Giacomo Mulas

unread,
Sep 30, 2021, 3:10:03 AM9/30/21
to
On Wed, 29 Sep 2021, Debian Bug Tracking System wrote:

> #995322: ipp-usb: prevents hp-toolbox and libsane-hpaio from working
>
> It has been closed by Brian Potkin <clarem...@gmail.com>.
>
> Much of your time would have been saved by reading the Release
> Notes. Please submit another report if you feel the documentation
> is inadequate.

Indeed it is. As I said, ipp-usb got automatically installed upon upgrading
to bullseye, together with tons of other packages. The only piece of
information I was shown was the NEWS.Debian file of ipp-usb (together with
dozens of other messages). The only thing it says is
"Existing or newly created queues on a USB connection for IPP-over-USB
capable devices using vendor drivers will not work while the ipp-usb
service is activated and managing the connection."
This is only about printing, no clue about other functions of the same
device. Only _after_ discovering that ipp-usb was preventing xsane to
connect to the scanner I looked at
https://wiki.debian.org/CUPSDriverlessPrinting
and found this warning in there:

"Note that IPP-over-USB reserves the USB interface connection with the
printer/scanner exclusively for itself and communication with a
printer/scanner device by software that does not operate using the
IPP-over-USB protocol becomes impossible while ipp-usb is running. This is
a consequence of the design of USB communication. It is not a bug in
ipp-usb. (omissis) Communicating with a USB connected scanner via classic
SANE backends such as libsane-hpaio, sane-pixma or sane-epson2 also becomes
impossible with the ipp-usb daemon active and running."

Had I seen _this_ warning in the release notes, _then_ this would have got
some alarm ringing and I would not have wasted hours tracking this.

So, please: do include the above warning, in full, in the release notes of
ipp-usb.

Also, it would be helpful to add at least a "Suggests" to install
also the package sane-airscan along with it, since this will restore most
multifunction devices to properly work again as scanners when ipp-usb is
running.

Thanks, and sorry for the inconvenience.

Bye
Giacomo

--
_________________________________________________________________

Giacomo Mulas <giacom...@inaf.it>
_________________________________________________________________

INAF - Osservatorio Astronomico di Cagliari
via della scienza 5 - 09047 Selargius (CA)

tel. +39 070 71180255
mob. : +39 329 6603810
_________________________________________________________________

"When the storms are raging around you, stay right where you are"
(Freddy Mercury)
_________________________________________________________________

Brian Potkin

unread,
Sep 30, 2021, 10:30:03 AM9/30/21
to
On Thu 30 Sep 2021 at 08:57:13 +0200, Giacomo Mulas wrote:

> On Wed, 29 Sep 2021, Debian Bug Tracking System wrote:
>
> > #995322: ipp-usb: prevents hp-toolbox and libsane-hpaio from working
> >
> > It has been closed by Brian Potkin <clarem...@gmail.com>.
> >
> > Much of your time would have been saved by reading the Release
> > Notes. Please submit another report if you feel the documentation
> > is inadequate.
>
> Indeed it is. As I said, ipp-usb got automatically installed upon upgrading
> to bullseye, together with tons of other packages. The only piece of
> information I was shown was the NEWS.Debian file of ipp-usb (together with
> dozens of other messages). The only thing it says is "Existing or newly
> created queues on a USB connection for IPP-over-USB
> capable devices using vendor drivers will not work while the ipp-usb
> service is activated and managing the connection."
> This is only about printing, no clue about other functions of the same
> device.

ipp-usb is a recommended package for cups-daemon and libsane1. That
the NEWS file does not mention scanning is a fair point. Does the
addotion of a third paragrapgh allay your concern?

If so, I woud suggest subnitting a report against ipp-usb. The change
is not extensive and could make it into a bullseye point release.

-------------------------------------------------------------------

ipp-usb uses the IPP-over-USB protocol to allow the setting up of a
driverless print queue for most USB connected modern multi-function
and a few modern USB-only devices. The default is to auto-setup the
queue with cups-browsed.

Existing or newly created queues on a USB connection for IPP-over-USB
capable devices using vendor drivers will not work while the ipp-usb
service is activated and managing the connection.

The same protocol also allows discovery of modern scanner devices via
libsane1. Oncve again, vendor drivers will not work while the ipp-usb
service is activated and managing the connection.

Details are at

https://wiki.debian.org/CUPSDriverlessPrinting

-------------------------------------------------------------------

BTW, Xsane should have given you two backends to access. hpaio will
not work bit escl should. I guess that was not the case. That is a bug
in escl and why the wiko points to sane-airscan as an alternative.

> Only _after_ discovering that ipp-usb was preventing xsane to
> connect to the scanner I looked at
> https://wiki.debian.org/CUPSDriverlessPrinting
> and found this warning in there:
> "Note that IPP-over-USB reserves the USB interface connection with the
> printer/scanner exclusively for itself and communication with a
> printer/scanner device by software that does not operate using the
> IPP-over-USB protocol becomes impossible while ipp-usb is running. This is
> a consequence of the design of USB communication. It is not a bug in
> ipp-usb. (omissis) Communicating with a USB connected scanner via classic
> SANE backends such as libsane-hpaio, sane-pixma or sane-epson2 also becomes
> impossible with the ipp-usb daemon active and running."
>
> Had I seen _this_ warning in the release notes, _then_ this would have got
> some alarm ringing and I would not have wasted hours tracking this.
>
> So, please: do include the above warning, in full, in the release notes of
> ipp-usb.

I think that duplicating in the Release Notes what is on the wiki is not
the way to go, but you could submit a bug against the release-notes
pseudo-package.

> Also, it would be helpful to add at least a "Suggests" to install also the
> package sane-airscan along with it, since this will restore most
> multifunction devices to properly work again as scanners when ipp-usb is
> running.

That would be for the libsane1 maintainer to consider. Note that he has
already declined to make sane-airscan a Recommends:

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=971658

Cheers,

Brian.
0 new messages