[Mrtrix-discussion] streamtrack hanging

[Donald Tournier]

Hi Luis,

I have seen strange behaviour like this before, in my case it was due to the
fact that none of the tracks generated met the criteria for inclusion. The
progress status line is only refreshed when a track is accepted, so if no
tracks are accepted the program appears to hang. Your case seems stranger
though, if you say the tracking proceeds in some cases and hangs half-way
through. I honestly don't know what could cause that...

So I'm not sure what's going on in your case, but if it's due to no tracks
being accepted, I'd suggest you try setting -cutoff, -initcutoff,
-minlength, and -curvature all to zero, and see if you get anything. If your
problem is still present, and the tracking genuinely does sometimes produce
some tracks and then hangs, I'm at a loss to explain it. The only thing I
can think of is that you might be running out of storage space...? Not sure
what the program would do in this case, I've never had that problem...

Another remote possibility is some strange problem with linux handling your
external storage device, which I notice you seem to be using. Maybe have a
look at the system logs (e.g. "less /var/log/messages") might provide a
clue.

If none of my suggestions are correct, the only way forward might be for you
to send me the file and the relevant command, and I'll see if I can debug it
when/if I have a spare minute...

Cheers,

Donald.


On 6 October 2011 07:40, Luis Concha <lconcha at gmail.com> wrote:

> Hello.
>
> I am experiencing problems with streamtrack, as it hangs when tracking from
> certain seeds. Oddly, it happens randomly, sometimes finishing the tracking
> and sometimes hanging in an unexpected manner. My seeds are always placed on
> high anisotropy voxels and I have tried both image-based seeds or
> coordinate-based seeds. Both approaches have always included a single voxel
> for seeding.
>
> Please see below the [DEBUG] output. It hangs after launching the threads,
> with all of them working at 100% CPU, but it does not finish. Any ideas?
>
> Thanks very much.
>
> Luis Concha
>
> $  streamtrack -info -debug -number 1000 -maxnum 10000 \
>   -seed 0,4,-2,2 \
>   SD_PROB \
>
> /media/IOMEGA1TB/denoise/data3/sintetico-homo/mrtrix/results//clean/out_CSD4.nii
> \
>
>  /media/IOMEGA1TB/denoise/data3/sintetico-homo/mrtrix/results//clean/t_0,4,-2,2.tck
>
> streamtrack [DEBUG]: reading key/value file "/home/lconcha/.mrtrix.conf"...
> streamtrack [INFO]: opening image
> "/media/IOMEGA1TB/denoise/data3/sintetico-homo/mrtrix/results//clean/out_CSD4.nii"...
> streamtrack [DEBUG]: preparing file
> "/media/IOMEGA1TB/denoise/data3/sintetico-homo/mrtrix/results//clean/out_CSD4.nii"
> streamtrack [DEBUG]: file
> "/media/IOMEGA1TB/denoise/data3/sintetico-homo/mrtrix/results//clean/out_CSD4.nii"
> mapped at 0x7f3cafd18000, size 60000352 (read-only)
> streamtrack [DEBUG]: sanitising transformation matrix...
> streamtrack [DEBUG]: unmapping file
> "/media/IOMEGA1TB/denoise/data3/sintetico-homo/mrtrix/results//clean/out_CSD4.nii"
> streamtrack [DEBUG]: setting up image
> "/media/IOMEGA1TB/denoise/data3/sintetico-homo/mrtrix/results//clean/out_CSD4.nii"...
> streamtrack [DEBUG]: sanitising transformation matrix...
> streamtrack [DEBUG]: setting up data increments for
> "/media/IOMEGA1TB/denoise/data3/sintetico-homo/mrtrix/results//clean/out_CSD4.nii"...
> streamtrack [DEBUG]: data increments initialised with start = 0, stride = [
> 1 100 10000 1000000 ]
> streamtrack [DEBUG]: mapping image
> "/media/IOMEGA1TB/denoise/data3/sintetico-homo/mrtrix/results//clean/out_CSD4.nii"...
> streamtrack [DEBUG]: file
> "/media/IOMEGA1TB/denoise/data3/sintetico-homo/mrtrix/results//clean/out_CSD4.nii"
> mapped at 0x7f3cafd18000, size 60000352 (read-only)
> streamtrack [DEBUG]: data mapper for image
> "/media/IOMEGA1TB/denoise/data3/sintetico-homo/mrtrix/results//clean/out_CSD4.nii"
> mapped with segment size = 15000000 (optimised)
> streamtrack [INFO]: launching 8 threads
>
>
> ... hangs here.
>
> _______________________________________________
> Mrtrix-discussion mailing list
> Mrtrix-discussion at www.nitrc.org
> http://www.nitrc.org/mailman/listinfo/mrtrix-discussion
>
>


-- 
Jacques-Donald Tournier (PhD)
Brain Research Institute, Melbourne, Australia
Tel: +61 (0)3 9035 7033
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.nitrc.org/pipermail/mrtrix-discussion/attachments/20111006/b1e22684/attachment.html