Bug 571520
| Summary: | TIFF problems: crash on fail; changed MaxStripSize | ||||||
|---|---|---|---|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Jim Shipman <JimShip> | ||||
| Component: | ghostscript | Assignee: | Tim Waugh <twaugh> | ||||
| Status: | CLOSED ERRATA | QA Contact: | Fedora Extras Quality Assurance <extras-qa> | ||||
| Severity: | medium | Docs Contact: | |||||
| Priority: | low | ||||||
| Version: | 12 | CC: | arne_woerner, twaugh | ||||
| Target Milestone: | --- | ||||||
| Target Release: | --- | ||||||
| Hardware: | x86_64 | ||||||
| OS: | Linux | ||||||
| Whiteboard: | abrt_hash:e77d5b4daa1f2d80f1327b19016b59c8161663f1 | ||||||
| Fixed In Version: | ghostscript-8.71-6.fc11 | Doc Type: | Bug Fix | ||||
| Doc Text: | Story Points: | --- | |||||
| Clone Of: | Environment: | ||||||
| Last Closed: | 2010-04-13 01:55:59 UTC | Type: | --- | ||||
| Regression: | --- | Mount Type: | --- | ||||
| Documentation: | --- | CRM: | |||||
| Verified Versions: | Category: | --- | |||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||
| Embargoed: | |||||||
| Attachments: |
|
||||||
|
Description
Jim Shipman
2010-03-08 18:24:08 UTC
Created attachment 398577 [details]
File: backtrace
The tiffg3 ghostscript device is missing a NULL check before calling TIFFCleanup(). However, this means that something else went wrong first. I've built a ghostscript package fixing this segfault. Could you please try it and tell me if/how it fails now? Thanks. http://koji.fedoraproject.org/koji/buildinfo?buildID=161482 hylafax still (i installed ghostscript.x86_64 8.71-4.1.fc12) doesn't work... -arne I expected it not to work, but could you please provide some information on exactly *how* it fails? Do you get an error message, is the fax image incorrect, is a fax sent at all, are there error logs somewhere, can you get debugging information out of it, etc? hylafax acts as if it sent everything correctly, although there were just 5 frames per page (that is for the tag line, i guess)... the kernel did not even log a segfault, also it seems to do it (at least when npviewer.bin segfault-s)... -arne i am quite sure that ghostscript causes the problem, because: when i do this: gs -sPaperSize=A4 -dBATCH -dNOPAUSE -sOutputFile=bla.tif -sDEVICE=tiffg3 ~/deploy/docs/... (without any problem or error) and then this: tiffcp -c g3:1d:fill -r 10000 bla.tif blah.tif and then this: sendfax -n -G -d "+49 30 ..." blah.tif everything behaves like before... *wag tail* -arne i hacked hylafax' bin/pdf2fax.gs script... this is the command that hylafax uses: /usr/bin/gs -q -sDEVICE=tiffg4 -dNOPAUSE -dSAFER=true -sPAPERSIZE=a4 -dFIXEDMEDIA -dBATCH -r209.10x196 -sOutputFile=docq/doc68.pdf;c1 docq/doc68.pdf.62 the resulting file looks good... when i force the bin/pdf2fax.gs script to use tiffg3 as -sDEVICE, then everything works fine again... so the tiffg4 device does something nasty/modern/unexpected possibly? -arne These problems for me happened after I updated my Fedora 12 system. efax-gtk broke causing the problems on my system. I rebuilt efax-gtk adding "parms.push_back("-dMaxStripSize=0");" at line 304 in efax-gtk-3.0.20/src/efax_controller.cpp. Everything now works fine on my system.
Jim Shipman
*w00t* that "-dMaxStripSize=0"-trick works for the bin/pdf2fax.gs script (in that call in the end of the script), too... -arne The default MaxStripSize had changed from 0 in ghostscirpt-8.70 to 8192 in ghostscript-8.71. I've set it back to 0 in ghostscript-8.71-5.fc12. Could you give it a go please, without the pdf2fax.gs/efax-gtk changes? http://koji.fedoraproject.org/koji/buildinfo?buildID=161952 i just sent a fax with my unpatched hylafax and the amd64 package from http://koji.fedoraproject.org/koji/buildinfo?buildID=161952 and hylafax sent 46 frames for one page, which is much more than just the tagline... looks good to me... -arne ghostscript-8.71-6.fc11 has been submitted as an update for Fedora 11. http://admin.fedoraproject.org/updates/ghostscript-8.71-6.fc11 ghostscript-8.71-6.fc12 has been submitted as an update for Fedora 12. http://admin.fedoraproject.org/updates/ghostscript-8.71-6.fc12 ghostscript-8.71-9.fc13 has been submitted as an update for Fedora 13. http://admin.fedoraproject.org/updates/ghostscript-8.71-9.fc13 The upstream fix was to change it to 1Mb so in ghostscript-8.71-6.fc12 I've switch to that. Could you please re-test just to make sure? but what if, that size increases to more than 1Mb? will there be again silent empty pages? or is it impossible to make a tiff file of that size and that resolution? -arne No single TIFF page can be more than that size, and there is one strip per page. ghostscript-8.71-6.fc12 has been pushed to the Fedora 12 testing repository. If problems still persist, please make note of it in this bug report. If you want to test the update, you can install it with su -c 'yum --enablerepo=updates-testing update ghostscript'. You can provide feedback for this update here: http://admin.fedoraproject.org/updates/ghostscript-8.71-6.fc12 ghostscript-8.71-6.fc11 has been pushed to the Fedora 11 testing repository. If problems still persist, please make note of it in this bug report. If you want to test the update, you can install it with su -c 'yum --enablerepo=updates-testing update ghostscript'. You can provide feedback for this update here: http://admin.fedoraproject.org/updates/ghostscript-8.71-6.fc11 ghostscript-8.71-9.fc13 has been pushed to the Fedora 13 stable repository. If problems still persist, please make note of it in this bug report. ghostscript-8.71-6.fc12 has been pushed to the Fedora 12 stable repository. If problems still persist, please make note of it in this bug report. ghostscript-8.71-6.fc11 has been pushed to the Fedora 11 stable repository. If problems still persist, please make note of it in this bug report. |