Bug 25363

Summary:	Printing on HP 3030 MF is taking forever with recommended driver (HP 3020 3030 Postscript)
Product:	Mageia	Reporter:	Frank Griffin <ftg>
Component:	RPM Packages	Assignee:	All Packagers <pkg-bugs>
Status:	NEW ---	QA Contact:
Severity:	normal
Priority:	Normal	CC:	mageia
Version:	Cauldron
Target Milestone:	---
Hardware:	All
OS:	Linux
Whiteboard:
Source RPM:	ghostscript	CVE:
Status comment:

Description Frank Griffin 2019-08-26 21:02:55 CEST

In a current cauldron fresh install, I add the HP 3030 MF printer I've been using for years using s-c-p.  This seemed to work fine, with two differences from historical behavior.  First, the URL chosen by s-c-p is an "hp:" URL.  Second, s-c-p is picking the driver on its own without giving me the usual choices.

I then tried to print a 5-page coupon.  The LED on the printer started blinking immediately, indicating that data was being received by the printer.  However, the first page took between 5 and 10 minutes to begin printing.  I left the house to do some errands, and when I got back an hour and a half later the second page still hadn't printed, although the LED was still blinking.

I see the same inordinate delay in printing a test page, so it's not specific to a particular application.

Apparently, whatever driver is being chosen by s-c-p is generating incredible amounts of printer data for jobs.

I've seen behavior similar to this in the past, although nowhere near as bad, and I've solved it in the past by choosing a different driver.  But that doesn't seem possible in s-c-p any more.

Any suggestions ?

Comment 1 Lewis Smith 2019-08-26 21:21:04 CEST

Thank you Frank for the report.
I have often had problems in the past with print jobs being queued but never printed. I doubt that in this case mega amounts of data are the cause. The only (hardly satisfactory) solution I found was to remove and re-install the printer - which did bring it back to life.

Assigning to neoclust as the system-config-printer maintainer.

Assignee: bugsquad => mageia

Comment 2 Frank Griffin 2019-08-26 21:36:36 CEST

Hi Lewis.  I've tried deleting and re-adding the printer several times, to no effect.  And the problem can't be jobs stuck in the queue, because the printer *is* receiving data.

"Stuck in the queue" usually happens because the driver crashes or errors out while trying to generate the native printer data.

There are probably two issues here.  The first is that whatever driver is being used *seems* to be sending mega amounts of data.  The second is that s-c-p is no longer allowing a choice of drivers.

The next thing I'll try is deleting and adding the printer back with CUPS.

Comment 3 Frank Griffin 2019-08-26 22:41:03 CEST

CUPS allowed me to define the printer with a "usb:" URL rather than "hp:", but the performance is the same.

Comment 4 Frank Griffin 2019-08-28 15:12:09 CEST

Changing this from s-c-p to ghostscript since the problem appears to be in the driver.  I've opened bug#25368 for the s-c-p aspect.

Summary: Printing on HP 3030 MF is taking forever => Printing on HP 3030 MF is taking forever with recommended driver (HP 3020 3030 Postscript)
Source RPM: system-config-printer => ghostscript

Comment 5 Frank Griffin 2019-08-28 18:20:49 CEST

The problem is the postscript drivers in ghostscript (or maybe CUPS, I'm not sure where they live).  I changed the driver manually with CUPS to "LaserJet 3030 pcl3, hpcups 3.19.6", and the printer works fine now.

Postscript should be fixed, or failing that, pcl3 made the "recommended" driver until it is.

Comment 6 Lewis Smith 2019-08-28 21:52:50 CEST

Thank you Frank for all your detective work. My c1 was somewhat naive in your case...

Changing assignee to global for Ghostscript, CC'ing blino as a past committer.

CC: (none) => mageia
Assignee: mageia => pkg-bugs