Debian Bug report logs - #815807
cups: printing on printers broadcast by old CUPS versions fails

Toggle useless messages

Message #5 received at submit@bugs.debian.org (full text, mbox, reply):

From: Dominik George <nik@naturalnet.de>

To: Debian Bug Tracking System <submit@bugs.debian.org>

Subject: cups: printing on printers broadcast by old CUPS versions fails

Date: Wed, 24 Feb 2016 16:57:41 +0100

Package: cups
Version: 2.1.3-1
Severity: normal

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Printing on printers published by CUPS 1.7.5 on a Debian stable print
server fails with the following:

 No suitable destination host found by cups-browsed.

The printers are recorded as being on „luna.local“, which is correct,
and I can resolve this name via mDNS and reach the host.

This is reproducible with several different print servers and clients,
always in the combination 2.1.3 client and 1.7.5 server.

- -- System Information:
Debian Release: stretch/sid
  APT prefers unstable
  APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable')
Architecture: amd64 (x86_64)

Kernel: Linux 4.4.0-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages cups depends on:
ii  cups-client            2.1.3-1
ii  cups-common            2.1.3-1
ii  cups-core-drivers      2.1.3-1
ii  cups-daemon            2.1.3-1
ii  cups-filters           1.8.2-2
ii  cups-ppdc              2.1.3-1
ii  cups-server-common     2.1.3-1
ii  debconf [debconf-2.0]  1.5.58
ii  ghostscript            9.18~dfsg-4
ii  libavahi-client3       0.6.32~rc+dfsg-1
ii  libavahi-common3       0.6.32~rc+dfsg-1
ii  libc-bin               2.21-9
ii  libc6                  2.21-9
ii  libcups2               2.1.3-1
ii  libcupscgi1            2.1.3-1
ii  libcupsimage2          2.1.3-1
ii  libcupsmime1           2.1.3-1
ii  libcupsppdc1           2.1.3-1
ii  libgcc1                1:5.3.1-9
ii  libstdc++6             5.3.1-9
ii  libusb-1.0-0           2:1.0.20-1
ii  lsb-base               9.20160110
ii  poppler-utils          0.38.0-2
ii  procps                 2:3.3.11-3

Versions of packages cups recommends:
ii  avahi-daemon                     0.6.32~rc+dfsg-1
ii  colord                           1.2.12-1
ii  cups-filters [ghostscript-cups]  1.8.2-2
ii  printer-driver-gutenprint        5.2.11-1

Versions of packages cups suggests:
ii  cups-bsd                                   2.1.3-1
pn  cups-pdf                                   <none>
ii  foomatic-db-compressed-ppds [foomatic-db]  20160212-1
ii  hplip                                      3.16.2+repack0-3
ii  printer-driver-hpcups                      3.16.2+repack0-3
pn  smbclient                                  <none>
ii  udev                                       229-1

- -- debconf information:
  cupsys/raw-print: true
  cupsys/backend: lpd, socket, usb, snmp, dnssd

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
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=oHy8
-----END PGP SIGNATURE-----

Message #10 received at 815807@bugs.debian.org (full text, mbox, reply):

From: Brian Potkin <claremont102@gmail.com>

To: Dominik George <nik@naturalnet.de>, 815807@bugs.debian.org

Subject: Re: Bug#815807: cups: printing on printers broadcast by old CUPS versions fails

Date: Thu, 25 Feb 2016 20:30:02 +0000

On Wed 24 Feb 2016 at 16:57:41 +0100, Dominik George wrote:

> Printing on printers published by CUPS 1.7.5 on a Debian stable print
> server fails with the following:
> 
>  No suitable destination host found by cups-browsed.

How do you receive this error message? In a log?

> The printers are recorded as being on „luna.local“, which is correct,
> and I can resolve this name via mDNS and reach the host.
> 
> This is reproducible with several different print servers and clients,
> always in the combination 2.1.3 client and 1.7.5 server.

Can you say a bit more about your setup? The message is associated with
backend/implicitclass.c of cups-filters. It implies remote queues are
disabled:

   if (!strcmp(dest_host, "NO_DEST_FOUND")) {
         /* All remote queues are either disabled or not accepting jobs, let
        CUPS retry after the usual interval */
        fprintf(stderr, "ERROR: No suitable destination host found by cups-browsed.\n");

Anything you can provide which would make the behaviour reproducible
would be helpful.

Regards,

Brian.

Message #15 received at 815807@bugs.debian.org (full text, mbox, reply):

From: Brian Potkin <claremont102@gmail.com>

To: 815807@bugs.debian.org

Cc: Dominik George <nik@naturalnet.de>, control@bugs.debian.org

Subject: Re: Bug#815807: cups: printing on printers broadcast by old CUPS versions fails

Date: Mon, 29 Feb 2016 19:12:27 +0000

tags 815807 moreinfo
thanks



On Thu 25 Feb 2016 at 20:30:02 +0000, Brian Potkin wrote:

> Anything you can provide which would make the behaviour reproducible
> would be helpful.

Any additional information which would progress this report?

Regards,

Brian.

Message #22 received at 815807-done@bugs.debian.org (full text, mbox, reply):

From: Brian Potkin <claremont102@gmail.com>

To: 815807-done@bugs.debian.org

Subject: Re: Bug#815807: cups: printing on printers broadcast by old CUPS versions fails

Date: Fri, 29 Apr 2016 17:51:45 +0100

On Wed 24 Feb 2016 at 16:57:41 +0100, Dominik George wrote:

> Printing on printers published by CUPS 1.7.5 on a Debian stable print
> server fails with the following:
> 
>  No suitable destination host found by cups-browsed.
> 
> The printers are recorded as being on „luna.local“, which is correct,
> and I can resolve this name via mDNS and reach the host.
> 
> This is reproducible with several different print servers and clients,
> always in the combination 2.1.3 client and 1.7.5 server.

Dear Debian User,

Further information on your issue has been requested twice but there has
been no response from you. Use of our limited, volunteer supported
resources is best served by not keeping open inactive bugs any longer
than desirable.

You may, of course, wish to review the reported problem. It would be
appreciated if any future investigation could be conducted using the
testing or unstable distributions and sufficient information to
reproduce it be provided. If considered necessary, this report could be
re-opened or a new bug report, possibly referencing this one, submitted.

Thank you for your report.

Regards,

Brian.

Message #35 received at 815807@bugs.debian.org (full text, mbox, reply):

From: Patrick Welche <prlw1@cam.ac.uk>

To: 815807@bugs.debian.org

Cc: prlw1@cam.ac.uk

Subject: cups: printing on printers broadcast by old CUPS versions fails

Date: Fri, 19 Aug 2016 12:42:51 +0100

Package: cups
Version: 2.1.3-1
Severity: normal

I just reproduced this, admittedly between ubuntu rather than vanilla debian
boxen, so it sounds like a real cups problem.

"Old" cups is 1.4.3
"New" cups is 2.1.3

extract of /etc/cups/printers.conf on old:

<Printer ufs2>
DeviceURI socket://...
...
Accepting Yes
Shared Yes

on new, presumably having cups-browsed it:

<Printer ufs2>
DeviceURI implicitclass:ufs2
...
Accepting Yes
Shared No


So, apparently no longer shared... (is that right?)

  Rejecting job because "ufs2" is not shared

quick round of lpadmin later, changed
to Shared Yes, and thereafter

held since "No suitable destination host found by cups-browsed."

Cheers,

Patrick

Message #40 received at 815807@bugs.debian.org (full text, mbox, reply):

From: Brian Potkin <claremont102@gmail.com>

To: Patrick Welche <prlw1@cam.ac.uk>, 815807@bugs.debian.org

Cc: control@bugs.debian.org

Subject: Re: Bug#815807: cups: printing on printers broadcast by old CUPS versions fails

Date: Mon, 22 Aug 2016 19:36:32 +0100

Reassign 815807 cups-browsed
thanks



Thank you for this additional information, Patrick.


On Fri 19 Aug 2016 at 12:42:51 +0100, Patrick Welche wrote:

> I just reproduced this, admittedly between ubuntu rather than vanilla debian
> boxen, so it sounds like a real cups problem.

The work involved in testing with Ubuntu installs and an ancient version
of CUPS didn't exactly thrill me. So I used two servers, A and B.

A has cups 1.7.5-11+deb8u1 (Jessie) and B cups 2.1.2-2. The client is an
up-to-date unstable.

> "Old" cups is 1.4.3
> "New" cups is 2.1.3
> 
> extract of /etc/cups/printers.conf on old:
> 
> <Printer ufs2>
> DeviceURI socket://...
> ...
> Accepting Yes
> Shared Yes
> 
> on new, presumably having cups-browsed it:
> 
> <Printer ufs2>
> DeviceURI implicitclass:ufs2

You have duplicate queues being broadcast on the network; is this your
intention?  cups-browsed combines them into one Implicit Class queue.

> ...
> Accepting Yes
> Shared No
> 
> 
> So, apparently no longer shared... (is that right?)

Not shared by the client is what it means, I think.

>   Rejecting job because "ufs2" is not shared
> 
> quick round of lpadmin later, changed
> to Shared Yes, and thereafter

I'm not very sure I appreciate what you did here and whether it is
important. What command? Which machine was it executed on.

Servers A and B were set up with identical print queues. A largish file
was sent five times in rapid succession from the client. I expected the
printing to be distributed between A and B because of load balancing.
It wasn't; everything went to A.

I then disabled the queue on A, expecting B to be used. Then came

 No suitable destination host found by cups-browsed.

in 'lpstat -t'.

Maybe I am doing some wrong but it does look like a bug in cups-browsed.

You will have to explain what you think is happening at your end and
whether it fits our experiences.

Regards,

Brian.

Message #49 received at 815807@bugs.debian.org (full text, mbox, reply):

From: Patrick Welche <prlw1@cam.ac.uk>

To: Brian Potkin <claremont102@gmail.com>

Cc: prlw1@cam.ac.uk

Subject: cups printing bug

Date: Mon, 22 Aug 2016 20:29:51 +0100

I'm afraid I'm just off on a singing trip for the week - I'll be near those
servers again next Tuesday - thanks for the follow up!

Cheers,

Patrick

Message #54 received at 815807@bugs.debian.org (full text, mbox, reply):

From: Brian Potkin <claremont102@gmail.com>

To: 815807@bugs.debian.org

Cc: Patrick Welche <prlw1@cam.ac.uk>

Subject: Re: Bug#815807: cups: printing on printers broadcast by old CUPS versions fails

Date: Sun, 28 Aug 2016 14:51:42 +0100

On Mon 22 Aug 2016 at 19:36:32 +0100, Brian Potkin wrote:

> On Fri 19 Aug 2016 at 12:42:51 +0100, Patrick Welche wrote:
> > 
> > <Printer ufs2>
> > DeviceURI implicitclass:ufs2
> 
> You have duplicate queues being broadcast on the network; is this your
> intention?  cups-browsed combines them into one Implicit Class queue.

That could be misleading. The duplicate queues are duplicate in the
sense they have identical names. At least two queues called ufs2 should
be on the network for cups-browsed to use the implicitclass URI. ufs2
and UFS2 are duplicates because queue names are case-insensitive when it
comes to printing.
 
> Servers A and B were set up with identical print queues. A largish file
> was sent five times in rapid succession from the client. I expected the
> printing to be distributed between A and B because of load balancing.
> It wasn't; everything went to A.
> 
> I then disabled the queue on A, expecting B to be used. Then came
> 
>  No suitable destination host found by cups-browsed.
> 
> in 'lpstat -t'.
> 
> Maybe I am doing some wrong but it does look like a bug in cups-browsed.

Let's go with "Maybe I am doing some wrong". It took me some time to
realise having cups-browsed running on A and B wasn't exactly conducive
to having two broadcast queues. Then I (perhaps rashly) purged and
reinstalled all cups related packages. Since then I have been unable to
reproduce the previous behaviour; load balancing works as described when
A and B are doing Bonjour broadcasting. (I realise this is not your
situation but did not want my initial remarks to muddy the waters).

Cheers,

Brian.

Message #59 received at 815807@bugs.debian.org (full text, mbox, reply):

From: Brian Potkin <claremont102@gmail.com>

To: 815807@bugs.debian.org

Cc: Patrick Welche <prlw1@cam.ac.uk>, control@bugs.debian.org

Subject: Re: Bug#815807: cups: printing on printers broadcast by old CUPS versions fails

Date: Sun, 4 Sep 2016 19:23:53 +0100

tags 815807 moreinfo
thanks


On Sun 28 Aug 2016 at 14:51:42 +0100, Brian Potkin wrote:

> On Mon 22 Aug 2016 at 19:36:32 +0100, Brian Potkin wrote:
> 
> > You have duplicate queues being broadcast on the network; is this your
> > intention?  cups-browsed combines them into one Implicit Class queue.

Have you managed to identify these queues so that we can make progress
with this report?

Regards,

Brian.

Message #64 received at 815807@bugs.debian.org (full text, mbox, reply):

From: Brian Potkin <claremont102@gmail.com>

To: 815807@bugs.debian.org

Cc: Patrick Welche <prlw1@cam.ac.uk>

Subject: Re: Bug#815807: cups: printing on printers broadcast by old CUPS versions fails

Date: Wed, 7 Sep 2016 19:52:11 +0100

On Sun 04 Sep 2016 at 19:23:53 +0100, Brian Potkin wrote:

> tags 815807 moreinfo
> thanks
> 
> 
> On Sun 28 Aug 2016 at 14:51:42 +0100, Brian Potkin wrote:
> 
> > On Mon 22 Aug 2016 at 19:36:32 +0100, Brian Potkin wrote:
> > 
> > > You have duplicate queues being broadcast on the network; is this your
> > > intention?  cups-browsed combines them into one Implicit Class queue.
> 
> Have you managed to identify these queues so that we can make progress
> with this report?

Enquiring minds await your response.

-- 
Brian

Message #69 received at 815807-done@bugs.debian.org (full text, mbox, reply):

From: Brian Potkin <claremont102@gmail.com>

To: 815807-done@bugs.debian.org

Subject: Re: Bug#815807: cups: printing on printers broadcast by old CUPS versions fails

Date: Mon, 17 Oct 2016 20:34:15 +0100

On Wed 07 Sep 2016 at 19:52:11 +0100, Brian Potkin wrote:

> On Sun 04 Sep 2016 at 19:23:53 +0100, Brian Potkin wrote:
> 
> > tags 815807 moreinfo
> > thanks
> > 
> > 
> > On Sun 28 Aug 2016 at 14:51:42 +0100, Brian Potkin wrote:
> > 
> > > On Mon 22 Aug 2016 at 19:36:32 +0100, Brian Potkin wrote:
> > > 
> > > > You have duplicate queues being broadcast on the network; is this your
> > > > intention?  cups-browsed combines them into one Implicit Class queue.
> > 
> > Have you managed to identify these queues so that we can make progress
> > with this report?
> 
> Enquiring minds await your response.

Four requests for information to progress this report. Zero reponse
of any substance.

Closing.

-- 
Brian.

Send a report that this bug log contains spam.