|
Bugzilla – Full Text Bug Listing |
| Summary: | EMU: ghostscript 9.27 breaks printing | ||
|---|---|---|---|
| Product: | [openSUSE] openSUSE Distribution | Reporter: | Luigi Baldoni <aloisio> |
| Component: | Printing | Assignee: | Johannes Meixner <jsmeix> |
| Status: | RESOLVED FIXED | QA Contact: | Johannes Meixner <jsmeix> |
| Severity: | Normal | ||
| Priority: | P5 - None | CC: | davidjmurdoch, emu, forgotten_7645792743, henk.vanvelden, ismail, jdelvare, maint-coord, Manfred.Haertel, martin.wilck, me, tchvatal, vcizek, werner |
| Version: | Leap 15.1 | ||
| Target Milestone: | --- | ||
| Hardware: | x86-64 | ||
| OS: | Other | ||
| URL: | https://paste.opensuse.org/52940726 | ||
| Whiteboard: | |||
| Found By: | --- | Services Priority: | |
| Business Priority: | Blocker: | --- | |
| Marketing QA Status: | --- | IT Deployment: | --- |
| Attachments: |
cups LogLevel debug file
cups-filters-297cc2de.patch -- commit 297cc2decc53e0b08eb03049fce375db33df2131 cups-filters-1.20.3-2.3.1.x86_64.rpm |
||
|
Description
Luigi Baldoni
2019-10-01 17:55:02 UTC
Same here on an USB connected HP Deskjet F4180 with HP Deskjet f4100 Series hpijs, 3.17.9. Downgrading helps. It is probably somewhere here: D [02/Oct/2019:12:35:12 +0200] [Job 115] GPL Ghostscript 9.27: Unrecoverable error, exit code 1 D [02/Oct/2019:12:35:12 +0200] [Job 115] Process is dying with \"Unable to determine number of pages, page count: -1 D [02/Oct/2019:12:35:12 +0200] [Job 115] \", exit stat 3 D [02/Oct/2019:12:35:12 +0200] [Job 115] Cleaning up... Same problem here on a HP LaserJet M1212nf. Downgrading worked around the problem. Adding Werner Fink who did the latest update. I am seeing the problem on both Leap 15.1 and SLED 12 SP4. I'm obviously available to provide debugging information or do more testing, I only need some guidance. Created attachment 820270 [details]
cups LogLevel debug file
Added log snippet.
I see this here frm log file bash> grep ^E Downloads/cups_gs927.log E [02/Oct/2019:16:35:36 +0200] Unable to open listen socket for address [v1.::1]:631 - Cannot assign requested address. E [02/Oct/2019:16:36:05 +0200] [CGI] cups-brf must be called as root E [02/Oct/2019:16:36:05 +0200] [cups-deviced] PID 20167 (cups-brf) stopped with status 1! E [02/Oct/2019:16:36:05 +0200] [cups-deviced] PID 20172 (gutenprint52+usb) stopped with status 1! E [02/Oct/2019:16:36:15 +0200] [Job 221] Job stopped due to filter errors; please consult the error_log file for details. the question rises if this cause the messages also included in the log file and seen in comment#2 > I see this here frm log file > E [02/Oct/2019:16:35:36 +0200] Unable to open listen socket for address [v1.::1]:631 - Cannot assign requested address. > E [02/Oct/2019:16:36:05 +0200] [CGI] cups-brf must be called as root > E [02/Oct/2019:16:36:05 +0200] [cups-deviced] PID 20167 (cups-brf) stopped with status 1! > E [02/Oct/2019:16:36:05 +0200] [cups-deviced] PID 20172 (gutenprint52+usb) stopped with status 1! My /var/log/cups/error_log shows all of the above since at least June 2018. > E [02/Oct/2019:16:36:15 +0200] [Job 221] Job stopped due to filter errors; please consult the error_log file for details. This instead has appeared only since the 1st of October 2019 when gutenprint 9.27 was pushed (and before setting the loglevel to debug). see also https://bugzilla.opensuse.org/show_bug.cgi?id=1131863#c7 Have the same problem here and https://bugzilla.opensuse.org/show_bug.cgi?id=1131863#c1 seems to have it seen coming... I see. I suppose the 9.27 update was rushed on account of the vulnerability. Same issue on Leap 15.0 as Leap 15.1. When previous ghostscript version 9.27-lp150.2.23.1 is installed to replace current version 9.26a-lp150.2.20.1 problem disappears. (In reply to David Murdoch from comment #10) > Same issue on Leap 15.0 as Leap 15.1. > > When previous ghostscript version 9.27-lp150.2.23.1 is installed to replace > current version 9.26a-lp150.2.20.1 problem disappears. @David: By updating the version field in the bug you *updated* it to 15.0 instead of *adding* it. Unfortunately bugzilla does not allow to select multiple versions. I changed it back to 15.1 as this is the most current version. (In reply to Martin Winter from comment #8) > see also https://bugzilla.opensuse.org/show_bug.cgi?id=1131863#c7 > > Have the same problem here and > https://bugzilla.opensuse.org/show_bug.cgi?id=1131863#c1 seems to have it > seen coming... Ahh ... thanks for this point!!! > As ghostscript 9.27 is now rolled-out on Leap 15.1, but cups-filters is still > on 1.20.3, printing is broken as already foreseen in comment #1. An update of > cups-filters is now absolutely necessary for Leap 15.1. this indeed I didn't know. I had tested only ghostscript for working with all available ps/pdf files an various printer drivers including cups! (In reply to Luigi Baldoni from comment #9) > I see. I suppose the 9.27 update was rushed on account of the vulnerability. sorry about, nevertheless I had tested ghostscript intensively but I simply had not known that versions of ghsotscript and cups-filters depend each other :( I'm currently building qpdf-9.0.1 and cups-filters-1.25.0 for Leap 15.0 and 15.1 on my home project https://build.opensuse.org/project/show/home:letsfindaway. I just copypac'ed the packages from Leap 15.2 and they built without causing any further dependency. As soon as they are published in the repository I will test them and come back here to inform you about the results. Perhaps backporting these packages from 15.2 could be an easy way to solve the problem. Builds are just finished! I installed these packages and the problem was gone. So we should really go in that direction. How could I trigger an official backport of those packages? What is the proper way to do this? Who should be contacted? Given that both packages are shared with SLE products, I guess the SUSE maintenance team should take care of that. Maintenance team, we'd need a fast maintenance update of qpdf and cups-filters in SLE 15, SLE 15 SP1 as well as SLE 12 SP4, to fix the regression caused by the recent maintenance update of ghostscript. I really wonder why both qpdf and cups-filters at build time do not depend on the version of libgs/ghostscript within the spec files. (In reply to Dr. Werner Fink from comment #16) > I really wonder why both qpdf and cups-filters at build time do not depend > on the version of libgs/ghostscript within the spec files. qpdf does not use libgs/ghostscript etc, it's a pretty standalone library. (In reply to Ismail Dönmez from comment #17) > (In reply to Dr. Werner Fink from comment #16) > > I really wonder why both qpdf and cups-filters at build time do not depend > > on the version of libgs/ghostscript within the spec files. > > qpdf does not use libgs/ghostscript etc, it's a pretty standalone library. Hmm .. strange, then I'd like to know why Martin dies have qpdf on his list ... beside cups-filters In cups-filters-1.25.0/INSTALL I found
**** IF YOU HAVE A NON-POSTSCRIPT PRINTER AND ARE NOT ****
**** RUNNING MAC OS X, YOU WILL ALSO NEED TO INSTALL GPL ****
**** GHOSTSCRIPT WITH THE "cups" DRIVER AFTER YOU INSTALL ****
**** CUPS AND BEFORE YOU INSTALL THIS PACKAGE. ****
and in cups-filters-1.25.0/NEWS
CHANGES IN V1.22.5
- foomatic-rip: Changed Ghostscript call to count pages in a
PDF file to use "runpdfbegin" and not the undocumented
Ghostscript internal "pdfdict", so that it works with
Ghostscript 9.27 and later (Debian bug #926576, Arch Linux
bug #62251).
... that seems to cause the trouble
``undocumented Ghostscript internal "pdfdict"''
Created attachment 820501 [details] cups-filters-297cc2de.patch -- commit 297cc2decc53e0b08eb03049fce375db33df2131 (In reply to Dr. Werner Fink from comment #19) > In cups-filters-1.25.0/INSTALL I found > > **** IF YOU HAVE A NON-POSTSCRIPT PRINTER AND ARE NOT **** > **** RUNNING MAC OS X, YOU WILL ALSO NEED TO INSTALL GPL **** > **** GHOSTSCRIPT WITH THE "cups" DRIVER AFTER YOU INSTALL **** > **** CUPS AND BEFORE YOU INSTALL THIS PACKAGE. **** > > and in cups-filters-1.25.0/NEWS > > CHANGES IN V1.22.5 > > - foomatic-rip: Changed Ghostscript call to count pages in a > PDF file to use "runpdfbegin" and not the undocumented > Ghostscript internal "pdfdict", so that it works with > Ghostscript 9.27 and later (Debian bug #926576, Arch Linux > bug #62251). > > ... that seems to cause the trouble > > ``undocumented Ghostscript internal "pdfdict"'' This seems the relevant part of cups-filters to get working cups back with ghostscript 9.27 SLE-15-SP2 is with latest cups-filters ... for SLE-12, SLE-12-SP3, and SLE-15 (which is also Leap 15.0 and Leap 15.1) I've added the commit as a patch and submitted this. Clearly this should be tested as well (In reply to Dr. Werner Fink from comment #16) > I really wonder why both qpdf and cups-filters at build time do not depend > on the version of libgs/ghostscript within the spec files. qpdf is on my list for a BuildRequires of cups-filters-1.25: BuildRequires: qpdf-devel >= 8.3.0 So it is not a runtime requirement but a build requirement. Anyway, when I install cups-filters-1.25, then also libqpdf.so.26 is required at runtime. (In reply to Martin Winter from comment #22) > (In reply to Dr. Werner Fink from comment #16) > > I really wonder why both qpdf and cups-filters at build time do not depend > > on the version of libgs/ghostscript within the spec files. > > qpdf is on my list for a BuildRequires of cups-filters-1.25: > > BuildRequires: qpdf-devel >= 8.3.0 > > So it is not a runtime requirement but a build requirement. Anyway, when I > install cups-filters-1.25, then also libqpdf.so.26 is required at runtime. ... and if you look at build.o.o https://build.opensuse.org/package/show/Printing/cups-filters you see that the builds for 15.0 and 15.1 fail because of a missing build dependency to qpdf-devel >= 8.3.0. That's why I also copied qpdf to my home project to resolve this dependency. Dr. Werner Fink wrote:
> SLE-15-SP2 is with latest cups-filters ... for SLE-12, SLE-12-SP3, and SLE-15 (which is also Leap 15.0 and Leap 15.1) I've added the commit as a patch and submitted this. Clearly this should be tested as well
Gladly. Is there a package somewhere or should I add that patch on my own?
Created attachment 820533 [details] cups-filters-1.20.3-2.3.1.x86_64.rpm (In reply to Luigi Baldoni from comment #25) > Dr. Werner Fink wrote: > > SLE-15-SP2 is with latest cups-filters ... for SLE-12, SLE-12-SP3, and SLE-15 (which is also Leap 15.0 and Leap 15.1) I've added the commit as a patch and submitted this. Clearly this should be tested as well > > Gladly. Is there a package somewhere or should I add that patch on my own? From internal build server ... I guess no ... nevertheless the build has provided for SUSE_SLE-15_Update/x86_64 the attached rpm. > From internal build server ... I guess no ... nevertheless the build has provided for SUSE_SLE-15_Update/x86_64 the attached rpm.
Appears to be working for me.
(In reply to Luigi Baldoni from comment #28) > > From internal build server ... I guess no ... nevertheless the build has provided for SUSE_SLE-15_Update/x86_64 the attached rpm. > > Appears to be working for me. Thanks a lot for this feedback! clear needinfo request, hope I supplied enough info in #22 and #23 SUSE-RU-2019:2567-1: An update that has one recommended fix can now be installed. Category: recommended (important) Bug References: 1152720 CVE References: Sources used: SUSE OpenStack Cloud Crowbar 8 (src): cups-filters-1.0.58-19.5.1 SUSE OpenStack Cloud 8 (src): cups-filters-1.0.58-19.5.1 SUSE Linux Enterprise Server for SAP 12-SP3 (src): cups-filters-1.0.58-19.5.1 SUSE Linux Enterprise Server 12-SP5 (src): cups-filters-1.0.58-19.5.1 SUSE Linux Enterprise Server 12-SP4 (src): cups-filters-1.0.58-19.5.1 SUSE Linux Enterprise Server 12-SP3-LTSS (src): cups-filters-1.0.58-19.5.1 SUSE Linux Enterprise Server 12-SP3-BCL (src): cups-filters-1.0.58-19.5.1 SUSE Linux Enterprise Desktop 12-SP5 (src): cups-filters-1.0.58-19.5.1 SUSE Linux Enterprise Desktop 12-SP4 (src): cups-filters-1.0.58-19.5.1 SUSE Enterprise Storage 5 (src): cups-filters-1.0.58-19.5.1 HPE Helion Openstack 8 (src): cups-filters-1.0.58-19.5.1 NOTE: This line indicates an update has been released for the listed product(s). At times this might be only a partial fix. If you have questions please reach out to maintenance coordination. SUSE-RU-2019:2566-1: An update that has one recommended fix can now be installed. Category: recommended (important) Bug References: 1152720 CVE References: Sources used: SUSE OpenStack Cloud 7 (src): cups-filters-1.0.58-15.5.1 SUSE Linux Enterprise Server for SAP 12-SP2 (src): cups-filters-1.0.58-15.5.1 SUSE Linux Enterprise Server for SAP 12-SP1 (src): cups-filters-1.0.58-15.5.1 SUSE Linux Enterprise Server 12-SP2-LTSS (src): cups-filters-1.0.58-15.5.1 SUSE Linux Enterprise Server 12-SP2-BCL (src): cups-filters-1.0.58-15.5.1 SUSE Linux Enterprise Server 12-SP1-LTSS (src): cups-filters-1.0.58-15.5.1 SUSE Enterprise Storage 4 (src): cups-filters-1.0.58-15.5.1 NOTE: This line indicates an update has been released for the listed product(s). At times this might be only a partial fix. If you have questions please reach out to maintenance coordination. The SLE Updates were released, Leap will follow shortly. SUSE-RU-2019:2568-1: An update that has one recommended fix can now be installed. Category: recommended (important) Bug References: 1152720 CVE References: Sources used: SUSE Linux Enterprise Module for Basesystem 15-SP1 (src): cups-filters-1.20.3-3.3.1 SUSE Linux Enterprise Module for Basesystem 15 (src): cups-filters-1.20.3-3.3.1 NOTE: This line indicates an update has been released for the listed product(s). At times this might be only a partial fix. If you have questions please reach out to maintenance coordination. openSUSE-RU-2019:2301-1: An update that has one recommended fix can now be installed. Category: recommended (important) Bug References: 1152720 CVE References: Sources used: openSUSE Leap 15.0 (src): cups-filters-1.20.3-lp150.2.3.1 openSUSE-RU-2019:2302-1: An update that has one recommended fix can now be installed. Category: recommended (important) Bug References: 1152720 CVE References: Sources used: openSUSE Leap 15.1 (src): cups-filters-1.20.3-lp151.3.3.1 *** Bug 1131771 has been marked as a duplicate of this bug. *** |