Freesurfers,
I am running freesurfer-Linux-centos5_x86_64-stable-pub-v5.0.0 on a Linux 2.6.18-164.11.1.el5_lustre.1.8.3 cluster and I have a group of subjects that consistently crash during the talairach transform with the following message in the recon-all.log:
ERROR: mpr2mni305 failed, see transforms/talairach_avi.log
The talairach_avi.log indicates that imgreg_4dfp abnormally terminates with a core dump:
Reading image: /lustre/scratch/jjoncaja/freesurfer/average/711-2C_as_mni_average_305.4dfp.img dimensions: 256 256 256 mmppix: 1.0000 -1.0000 -1.0000 center: 128.0000 -129.0000 -129.0000 talsrcimg_g11.4dfp.img 256 256 256 1 1.000000 1.000000 1.000000 orientation T byte_order littleendian Reading image: talsrcimg_g11.4dfp.img dimensions: 256 256 256 mmppix: 1.0000 -1.0000 -1.0000 center: 128.0000 -129.0000 -129.0000 Abort (core dumped) ERROR: 'imgreg_4dfp /lustre/scratch/jjoncaja/freesurfer/average/711-2C_as_mni_average_305 none talsrcimg_g11 none talsrcimg_to_711-2C_as_mni_average_305_t4 4355' failed! status=134 ERROR: mpr2mni305 execution aborted
This problem was replicated on FS 4.2, 4.3, and 5.0. While datasets I am analyzing were not exactly optimized for use with Freesurfer (screen shots attached), most of my subject either process okay or return error message identifying the problem. Before dropping the "core dump" subjects from the analysis I wanted to see what I can do to save them.
If needed I can drop a tarball containing a single problematic subject into the filedrop or email the 001.mgz whichever is more convenient. Thank you in advance for all your help.
Greetings,
Jacek Jonca-Jasinski ღ
^..^<
Jacek Jonca-Jasinski Texas Tech University
As we act, let us not become the evil that we deplore. --Barbara Lee
