Unable to print after libgs 9.26 installed

[Matt Walters]

I'm receiving the following error in the cups error_log after libgs was updated to 9.26:

D [09/May/2019:11:19:22 -0400] [Job 6927] Filetype: PDF
D [09/May/2019:11:19:22 -0400] [Job 6927] Storing temporary files in /tmp
D [09/May/2019:11:19:22 -0400] [Job 6927] PID 28996 (/usr/lib/cups/filter/pdftopdf) exited with no errors.
D [09/May/2019:11:19:22 -0400] [Job 6927] GPL Ghostscript 9.26: Unrecoverable error, exit code 1
D [09/May/2019:11:19:22 -0400] [Job 6927] Process is dying with \"Unable to determine number of pages, page count: -1
D [09/May/2019:11:19:22 -0400] [Job 6927] \", exit stat 3
D [09/May/2019:11:19:22 -0400] [Job 6927] Cleaning up...
D [09/May/2019:11:19:22 -0400] [Job 6927] PID 28997 (/usr/lib/cups/filter/foomatic-rip) stopped with status 3.
D [09/May/2019:11:19:22 -0400] [Job 6927] prnt/backend/hp.c 919: ERROR: null print job total=0
D [09/May/2019:11:19:22 -0400] [Job 6927] PID 28998 (/usr/lib/cups/backend/hp) exited with no errors

[Matt Walters (matt.walters)]

Some clarification. I'm using ubuntu 18.04 with an HP Laserjet 500 MFP 570dn... two servers, two of the same printer, and the same issue. It appears to have occured after the following unattended upgrade:

Upgrade: libgs9:amd64 (9.26~dfsg+0-0ubuntu0.18.04.8, 9.26~dfsg+0-0ubuntu0.18.04.9), ghostscript:amd64 (9.26~dfsg+0-0ubuntu0.18.04.8, 9.26~dfsg+0-0ubuntu0.18.04.9), ghostscript-x:amd64 (9.26~dfsg+0-0ubuntu0.18.04.8, 9.26~dfsg+0-0ubuntu0.18.04.9), libgs9-common:amd64 (9.26~dfsg+0-0ubuntu0.18.04.8, 9.26~dfsg+0-0ubuntu0.18.04.9)

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Matt Walters (matt.walters)]

installing cups-filter 1.20.2 resolved the issue. Thanks everyone!

[rtimai (rtimai)]

While the last post was in 05/2019, this Ubuntu page on launchpad is still relevant for Debian users, as conservative Debian is several versions behind most distros. I am encountering this compatibility issue between cups-filters 1.21.6-5 amd64 and ghostscript 9.27~dfsg-2+deb10u4 amd64 since 11/2020, i.e., since GS was upgraded (possibly a regression error, since 1.21 was supposed to have fixed a previous 'filters failed' error.)

Debian does not appear to support specifying downgraded package versions.

~$ sudo apt-get install ghostscript=9.26a~dfsg-0+deb9u1
Reading package lists... Done
Building dependency tree
Reading state information... Done
E: Version '9.26a~dfsg-0+deb9u1' for 'ghostscript' was not found
~$

This issue appears to cross multiple printer series -- I am using an HP-Deskjet-1112, and am unable to apply selective page printing for double-sided results on a single-function printer (e.g., printing ODD pages, flip-reloading and then printing EVEN pages.) This is for work schedule charts, so it's not trivial to me.

If anyone can provide a workaround, I'd greatly appreciate it if you can provide specific instructions. Thanks all, for reading.

[rtimai (rtimai)]

Forgot to mention, Debian 10 Buster currently installs
cups-filters 1.21.6-5 with ghostscript 9.27~dfsg-2+deb10u4

I tried and failed to revert the ghostscript version.

~$ lsb_release -ds ; uname -srm ; gnome-shell --version ; echo $XDG_SESSION_TYPE
Debian GNU/Linux 10 (buster)
Linux 4.19.0-13-amd64 x86_64
GNOME Shell 3.30.2
x11
~$

[rtimai (rtimai)]

Attempted the following suggested procedures, without success.

/var/log/apt/history.log
sudo apt reinstall libgs9-common
  Reinstall: libgs9-common:amd64 (9.27~dfsg-2+deb10u4)
sudo apt reinstall hpijs-ppds printer-driver-hpijs
  Reinstall: hpijs-ppds:amd64 (3.18.12+dfsg0-2), printer-driver-hpijs:amd64 (3.18.12+dfsg0-2)

On the chance that the "Filters failed" issue involved the PDF viewer Evince, I installed the PDF Viewer browser extension in Chromium, enabled local file access per instructions, and I am now able to print selected PDF pages without error. Interestingly, Evince continues to fail filtered print jobs, but now without returning an error message. The cups error log appears to mark the failed jobs as "completed" now.

My point here is that a potential workaround for this complex issue may be to try using an alternate PDF viewer.