Duke-UNC Brain Imaging and Analysis Center
BIAC Forums | Profile | Register | Active Topics | Members | Search | FAQ
Username:
Password:
Save Password   Forgot your Password?
 All Forums
 Support Forums
 Analysis Software Support
 FSL 4.1.4
 New Topic  Reply to Topic
 Printer Friendly
Author Previous Topic Topic Next Topic  

diaz
BIAC Alum

USA
212 Posts

Posted - Aug 18 2009 :  1:01:36 PM  Show Profile  Reply with Quote
Hi,
I'm having trouble getting FSL 4.1.4 to run on the cluster. The last error I get is FILM did not complete - probably ran out of memory. At the beginning of the log file, I see the following:


/usr/local/fsl-4.1.4-centos4_64/bin/feat_model design
Can't find key fmri(evs_vox)

mkdir .files;cp /usr/local/fsl-4.1.4-centos4_64/doc/fsl.css .files;ln -s /usr/local/fsl-4.1.4-centos4_64/doc/images .files/images
ln: creating symbolic link `.files/images' to `/usr/local/fsl-4.1.4-centos4_64/doc/images': Operation not supported

Any thoughts? Has anyone had luck running FSL 4.1.4 on the cluster?

Thanks!
Michele

Michele T. Diaz, Ph.D.
Associate Director
Brain Imaging and Analysis Center

dvsmith
Advanced Member

USA
218 Posts

Posted - Aug 18 2009 :  1:08:12 PM  Show Profile  Visit dvsmith's Homepage  Reply with Quote
Save for the issue I posted about last week, I haven't had any issues with FSL 4.1.4 on the cluster. However, Brandi had a similar error ("Can't find key").

Try making a new template with the new version. And, you should run feat using $FSLDIR/bin/feat. This should fix it.
https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind0808&L=FSL&D=0&P=182313

Cheers,
David
Go to Top of Page

petty
BIAC Staff

USA
453 Posts

Posted - Aug 18 2009 :  1:15:21 PM  Show Profile  Reply with Quote
did you make a new FSF design template with the newer version of Feat, or are you trying to run an older model through?
Go to Top of Page

diaz
BIAC Alum

USA
212 Posts

Posted - Aug 18 2009 :  1:21:54 PM  Show Profile  Reply with Quote
so that's the thing, I had tried to create a new *.fsf template.
I invoke, what I thought was the new version of FSL by typing this (at q interact)
/usr/local/fsl-4.1.4-centos4_64/bin/fsl and I've also tried this
/usr/local/fsl-4.1.4-centos4_64/bin/Feat

Now I'm realizing that it seems both are bringing up FSL 4.0, and the *.fsf file that both creates is missing the variable described above (set fmri(evs_vox) 0)

So I'm sure that using an old *.fsf is the problem, but now I'm not sure how to load FSL 4.1.4.

Michele T. Diaz, Ph.D.
Associate Director
Brain Imaging and Analysis Center
Go to Top of Page

gunes
BIAC Alum

45 Posts

Posted - Aug 18 2009 :  1:27:18 PM  Show Profile  Reply with Quote
If you add the following lines as Josh described earlier into your profile, you may solve your problem:

http://www.biac.duke.edu/forums/topic.asp?TOPIC_ID=1190&SearchTerms=fsl,4.1.4

Hope it helps.

Gunes
Go to Top of Page

diaz
BIAC Alum

USA
212 Posts

Posted - Aug 18 2009 :  1:29:55 PM  Show Profile  Reply with Quote
Thanks, I have those lines in my scripts, but I wasn't sure that I wanted to switch over to the new version completely. But I can just edit my .bash_profile file when I want to switch back and forth.

Thanks for everyone's help!
Michele

Michele T. Diaz, Ph.D.
Associate Director
Brain Imaging and Analysis Center
Go to Top of Page

wang
BIAC Alum

USA
57 Posts

Posted - Nov 10 2009 :  09:32:27 AM  Show Profile  Reply with Quote

I am using fsl 4.1.4 version to do a third level group analysis, and got some error messages as the following:

/usr/local/fsl/bin/fslmaths mean_func -Tmean mean_func
syntax error in expression "156.855189 + ": premature end of expression
while executing
"expr $mean_lcon + $mean_lcon_incr "
(procedure "feat5:proc_gfeatprep" line 272)
invoked from within
"feat5:proc_gfeatprep "
("-gfeatprep" arm line 2)
invoked from within
"switch -- [ lindex $argv $argindex ] {

-I {
incr argindex 1
set session [ lindex $argv $argindex ]
}

-D {
incr argindex 1
set..."
("for" body line 2)
invoked from within
"for { set argindex 1 } { $argindex < $argc } { incr argindex 1 } {
switch -- [ lindex $argv $argindex ] {

-I {
incr argindex 1
set ses..."
(file "/usr/local/fsl/bin/feat" line 137)

Does anyone know what's wrong with the mean_fun image? The mean_func images from the reg_standard directory look fine to me. My inputs are from the first level analysis since I only have one run per EV. Thanks for any suggestions.


lihong
Go to Top of Page

gunes
BIAC Alum

45 Posts

Posted - Nov 10 2009 :  09:54:44 AM  Show Profile  Reply with Quote
Hi Lihong,

Can you explain your design a little bit to help for your scripts? i have one design that has a few runs and each run has different EVs. Based on my knowledge i can say as follows:
If you have one run per each EV, At level 2 you have to combine all main EVs and create all contrasts that you are interested in using Level 3. Maybe you are missing Level 2 stat if you are using several runs with different EVs.

I hope this helps.

Gunes
Go to Top of Page

wang
BIAC Alum

USA
57 Posts

Posted - Nov 10 2009 :  2:28:11 PM  Show Profile  Reply with Quote

Thanks Gunes. For each EV I only have one run, so I don't need to run level2. The third level group analysis using results directly from level1 worked fine previously. However, I got this error message when I rerun the same model with minor changes.


lihong
Go to Top of Page

petty
BIAC Staff

USA
453 Posts

Posted - Nov 10 2009 :  2:56:49 PM  Show Profile  Reply with Quote
Lihong,

Did you happen to remove any subjects, or move any of their data?

Looks like during the mean-func creation fsl is expecting there to be more data than you actually have.

Go to Top of Page

wang
BIAC Alum

USA
57 Posts

Posted - Nov 10 2009 :  3:09:13 PM  Show Profile  Reply with Quote

Oh! Indeed I had to delete three subjects because they are lacking of behavior data. I have the same number of input when I define the model though. Anyway, this is really an important hint. I will double check the input and the model.

Thanks!

lihong
Go to Top of Page

simon
Starting Member

USA
0 Posts

Posted - Oct 17 2014 :  5:46:42 PM  Show Profile  Reply with Quote
Not sure if this is a related problem, but maybe. When i try to run models (off the GUI) from a remote laptop with a munin drive mounted i get the following error:

terminal output:

simons-air:TF006 simonwdavis$ ln: .files/images: Operation not supported
while executing
"fsl:exec "mkdir .files;cp ${FSLDIR}/doc/fsl.css .files;ln -s ${FSLDIR}/doc/images .files/images""
invoked from within
"if { $done_something == 0 } {

if { ! $fmri(inmelodic) } {
if { $fmri(level) == 1 } {
#{{{ FEAT first-level analysis

for { set session 1 } ..."
(file "/usr/local/fsl/bin/feat" line 207)

report_log.html:

/bin/cp /tmp/feat_hADbdk.fsf design.fsf /usr/local/fsl/bin/feat_model design mkdir .files;cp /usr/local/fsl/doc/fsl.css .files;ln -s /usr/local/fsl/doc/images .files/images ln: .files/images: Operation not supported FATAL ERROR ENCOUNTERED

Doesn't happen if the data is local. Is this a permissions issue? Qsub working and all that, but would be nice to test out models before batching.
Go to Top of Page

petty
BIAC Staff

USA
453 Posts

Posted - Oct 17 2014 :  6:16:05 PM  Show Profile  Reply with Quote
You can't make symbolic links on the cifs mounted file system. The "ln -s" is what is throwing the error.

You could, however, run this on the tmp drive
Go to Top of Page

syam.gadde
BIAC Staff

USA
421 Posts

Posted - Oct 19 2014 :  08:46:19 AM  Show Profile  Reply with Quote
We see the same errors in feat log files on the cluster but they are generally not fatal for some reason. If this is a Mac laptop it may have something to do with the shell that is running the commands under the hood. Even if they're both bash (the cluster default) there may be an option that tells feat or bash not to break on these errors (but which also means it will plunge ahead even in the presence of other errors which means you will have to search the log files for the first real error ).
Go to Top of Page
  Previous Topic Topic Next Topic  
 New Topic  Reply to Topic
 Printer Friendly
Jump To:
BIAC Forums © 2000-2010 Brain Imaging and Analysis Center Go To Top Of Page
This page was generated in 0.38 seconds. Snitz Forums 2000