abrt 1.1.1 detected a crash. architecture: i686 Attached file: backtrace cmdline: /usr/bin/xsane --device-settings /home/goudsmid/.sane/xsane/EpsonCalibrated component: xsane crash_function: e2_setup_block_mode executable: /usr/bin/xsane global_uuid: 312562977f3e83a61e61cb161d98b7dcf36e934d kernel: 2.6.33.5-124.fc13.i686.PAE package: xsane-0.997-8.fc13 rating: 4 reason: Process /usr/bin/xsane was killed by signal 8 (SIGFPE) release: Fedora release 13 (Goddard) How to reproduce ----- 1. Start xsane 2. Configure xsane to save to a jpeg file (if set to TEXT, this doesn't seem to happen) 3. Create a preview 4. Change the scan area by clicking and dragging in the preview area a couple of times => At some point xsane crashes
Created attachment 429857 [details] File: backtrace
Would you please attach the device settings file /home/goudsmid/.sane/xsane/EpsonCalibrated to this ticket? Please also check, if the issues remains if you move away your complete XSane settings, e.g. rename ~/.sane/xsane to something else, then check again. Thanks.
Package: xsane-0.997-8.fc13 Architecture: i686 OS Release: Fedora release 13 (Goddard) How to reproduce ----- 1. Same as my previous report, except I removed .sane directory first
Sorry, my previous comment was the result of Abrt. I tried again after I removed the .sane directory, and can crash xsane in pretty much the same way. It may be useful to know the crash only happens with my Epson Perfection 3200 Photo, and not with the scanner in my HP Color Laserjet CM2320. Do you want me to attach the new backtrace as well ?
Well, if it's significantly different it might be worthwhile. If you don't know, just attach it, I'll sort it out whether I need it or not ;-).
Created attachment 430469 [details] Backtrace from clean xsane run The above backtrace is from the clean run (with .sane moved out of the way). I'll let you decide if it contains additional valuable information. Thank you for analysing this. Xsane is rather important in our workflow here.
Created attachment 430545 [details] Config file used with xsane Attached is my EpsonCalibrated.drc file, which is used in when the first backtrace was generated. I noticed sane-backends comes with two possible backends for Epson scanners: epson and epson2. I tried both of them. Only epson2 is susceptible to the crash. Also, with the old .sane in place and using the configuration file, the crash seems to happen much more often. Usually the first or second time I set a scan area in the preview window, xsane bails out. With a fresh .sane directory, not using the custom configuration file, I usually have to set the scan area multiple times in a row before it crashes. This is not scientifically established though, just my impression. Hopefully that helps narrow down the bug location. Note: although the epson driver seems to work with my scanner, it doesn't give good scan results and the preview is dog slow. Fixing epson2 is definitely a better solution here. If there is anything more I can do to help you fix this, please let me know.
This bug has been fixed with the xsane updates a couple of days ago. These updates were to fix bug 671534, so I'll mark this one as a duplicate. *** This bug has been marked as a duplicate of bug 671534 ***