Bug 27260 - Choosing "rpmnew" for a recent update of sane-backends results in loss of scanner function of Officejet 6100
Summary: Choosing "rpmnew" for a recent update of sane-backends results in loss of sca...
Status: RESOLVED OLD
Alias: None
Product: Mageia
Classification: Unclassified
Component: RPM Packages (show other bugs)
Version: 7
Hardware: All Linux
Priority: Normal critical
Target Milestone: ---
Assignee: All Packagers
QA Contact:
URL:
Whiteboard:
Keywords: Triaged
Depends on:
Blocks:
 
Reported: 2020-09-08 05:45 CEST by Thomas Andrews
Modified: 2021-09-07 14:11 CEST (History)
4 users (show)

See Also:
Source RPM: sane-1.0.28-1.1.mga7.src.rpm
CVE:
Status comment:


Attachments

Description Thomas Andrews 2020-09-08 05:45:07 CEST
Description of problem: After getting the updated packages associated with Bug 26712, the user was offered the chance to choose to use "rpmnew" as the config for sane-backends, or to do nothing. On my 64-bit Plasma production install, I chose to use "rpmnew," and lost the scanner functions of my HP Officejet 6100. The printer continues to work. Simple Scan reports that a scanner driver is needed, and the Gimp's sane module simply says there are no devices. The scanner had functioned perfectly five days ago, before the update was installed.

On a 32-bit Xfce install on the same hardware that hadn't been updated in a long time, I (without thinking) again chose "rpmnew" with the same result. On yet another 64-bit Plasma install that also hadn't been updated in a while I choose "do nothing," and the scanner still functions perfectly.

The scanner is installed using system-config-printer with hplip at the same time as the printer, so I tried removing and re-installing the Officejet. No change. I did not try that with the install that still works, as I did not want to risk losing the use of the scanner.


Version-Release number of selected component (if applicable):
sane-backends-1.0.28-1.1.mga7

How reproducible: Seems to be every time. I do not know if it would happen with other HP scanners/all-in-ones. 

I'm going to label this as "Critical," because if affected the user loses all scanner function. If "Major" is better because it appears that not all hardware is affected, feel free to change it.
Comment 1 Aurelien Oudelet 2020-09-08 16:07:41 CEST
Hi, thanks for reporting this bug.

Do these commands do nothing?

# sane-find-scanner
# scanimage -L

Have you tried to restore config file from downloaded older version RPM in repository?

Keywords: (none) => Triaged

Comment 2 Nicolas Salguero 2020-09-08 18:42:59 CEST
Hi,

Can you try adding a line containing only "hpaio" at the end of /etc/sane.d/dll.conf and see if it solves the issue?

That line does not come from a package but is added by some config tools when configuring an HP scanner so restoring from an older RPM will not solve the problem.

Best regards,

Nico.

CC: (none) => nicolas.salguero

Comment 3 Lewis Smith 2020-09-08 20:58:06 CEST
It seems useful to know:
- what config file was changed by the update
 https://bugs.mageia.org/show_bug.cgi?id=26712#c6
- where the change came from.

Because NicolasS is already involved in this new (& previous) bug, assigning it to you (removed CC). CC'ing DavidG who did that update.

Assignee: bugsquad => nicolas.salguero
Source RPM: (none) => sane-1.0.28-1.1.mga7.src.rpm
CC: nicolas.salguero => geiger.david68210

Comment 4 Thomas Andrews 2020-09-09 02:01:56 CEST
(In reply to Nicolas Salguero from comment #2)
> Hi,
> 
> Can you try adding a line containing only "hpaio" at the end of
> /etc/sane.d/dll.conf and see if it solves the issue?
> 
> That line does not come from a package but is added by some config tools
> when configuring an HP scanner so restoring from an older RPM will not solve
> the problem.
> 
> Best regards,
> 
> Nico.

That took care of it.

So now, what's the best way to get this out to everyone else?
Rolf Pedersen 2020-09-09 04:14:46 CEST

CC: (none) => rolfpedersen

Comment 5 Rolf Pedersen 2020-09-09 04:34:51 CEST
Same thing happened to me with a couple of Brother MFC set up on one 64-bit MGA7 desktop:  when I chose to use rpmnew, scanners were no longer found by xsane (or simple scan or xscanimage).

We used to have to make udev rules for these Brother MFC's, so I tried one of those recipes without success, although it might have contributed to eventual success.  I also re-installed two versions of a Brother software, brscan2 and brscan4, for the one older and one newer machine.

Brother provides a script, e.g. linux-brprinter-installer-2.2.1-1, and re-running this script on each machine as the third swing of my hammer brought the scanners back into my world.  FWIW.
Comment 6 Nicolas Salguero 2020-09-09 09:57:48 CEST
(In reply to Thomas Andrews from comment #4)
> So now, what's the best way to get this out to everyone else?

I do not know as I am not a expert of sane.  All I know is that it is not a new issue but I already encountered that problem with Mandriva at least (I do not remember whether or not the problem was present in Mandrake): I always update from version N to version N+1 of Mageia and do not make a fresh install so I always have to take care of ".rpmnew"/".rpmsave" files.

Best regards,

Nico.

CC: (none) => nicolas.salguero
Assignee: nicolas.salguero => pkg-bugs

Comment 7 Thomas Andrews 2020-09-09 14:11:14 CEST
"That line does not come from a package but is added by some config tools when configuring an HP scanner so restoring from an older RPM will not solve the problem."

I actually suspected something of the sort when I encountered the bug, which is why the first thing I tried was to remove and re-install the Officejet. That didn't work, I suppose, because when you "remove" a printer everything isn't actually "removed." So when the printer/scanner was re-installed, sane was not reconfigured.



Back when I was less "experienced," I always chose "do nothing" over "rpmnew" because when I looked at the changes all I saw looked like so much gobbledegook to me. Doing nothing, rather than doing something I didn't understand, seemed the wiser choice.

But then I joined QA, and saw security update after security update. Reasoning I perhaps was not taking advantage of some of those security changes by doing nothing, I reversed my policy. And I ran into this.

I'm sure there must be several of our users who face the same dilemma when confronted by the rpmnew or nothing choice. We really should do what we can to avoid breaking their systems, whichever way they choose to go.
Comment 8 Aurelien Oudelet 2020-09-09 15:06:26 CEST
Sometimes .rpmnew files continain new settings or updated link/documentations.
But these new config files should not overwrite users/admin or local specific settings.

Instead of letting user in front of a dilemna choice, a readme file should be provided or permit the fusion of files as the GUI shows a diff between original and new file. Also the GUI could propose to edit file.
Comment 9 Dave Hodgins 2020-09-09 15:10:51 CEST
I normally choose the do nothing option, and then use kdiff3 to merge the
changes in the new file with any customizations I've made or a config tool
made in the existing file.

CC: (none) => davidwhodgins

Comment 10 Aurelien Oudelet 2021-07-06 13:17:51 CEST
Mageia 7 is EOL since July 1st 2021.
There will not have any further bugfix for this release.

You are encouraged to upgrade to Mageia 8 as soon as possible.

@reporter, if this bug still apply with Mageia 8, please let us know it.

@packager, if you work on the Mageia 7 version of your package, please check the Mageia 8 package if issue is also present. In this case, please fix the Mageia 8 version instead.

This bug report will be closed OLD if there is no further notice within 1st September 2021.
Comment 11 Marja Van Waes 2021-09-07 14:11:19 CEST
Hi bug reporter and hi assignee and others involved,

Please reopen this bug report if it is still valid for Mageia 8 or 9(cauldron), and change "Version:" in the upper left of this report accordingly.

This report is being closed as OLD because it was filed against Mageia 7, for which  support ended on June 30th 2021.

Thanks,
Marja

Resolution: (none) => OLD
Status: NEW => RESOLVED


Note You need to log in before you can comment on or make changes to this bug.