[Brains-users] Error message during measurement of T50 volumes
Eugene Zeien
eugene-zeien at uiowa.edu
Tue Mar 7 07:05:56 PST 2006
I was unable to find an occurrence of 'chown' in the current brains2 source code.
Unfortunately, there are still a couple of 'chgrp' commands that are dependent
upon the existence of group 'stdwkup' and the user's inclusion in that group.
Rather than use 'sudo' to run brains2, you can safely ignore the complaint.
Alternatively, you can edit scripts and change 'stdwkup' to 'office-users'.
Here's a one-liner that will change the default group for the fitting script.
sed -i".bak" 's/stdwkup/office-users/g' `find /opt/brains2 -name brains2_fit.csh`
[root at opteron ipl_config]# fgrep chgrp /opt/brains2/bin/brains2_fit.csh
chgrp stdwkup $ACPCdir/$tracer/fit_$airMode.log
chgrp stdwkup $airFile
chgrp stdwkup $ACPCdir/$tracer/check_fit.brains2
Since you are likely using an older version of brains2 than the source I'm searching,
find /opt/brains2 -type f -exec fgrep -l stdwkup '{}' \;
Will give you a list of the files which may contain references to this group.
I know there is one .tcl script that uses 'stdwkup' as a function-name, and you
probably don't want to edit the binaries.
Hope this helps,
gene
Simone Brockman wrote:
> Hi Eugene, thanks for the info - I changed the line as directed , but the problem still exists. The error message reflects the changed code but the operation still fails... our programming support guy has provided info about what he has done (see below)is there anything thing there that you can see could be causing the problem?
>
>
> This is the brains2 setup that is present on Laura: -
>
> - Running on an Ubuntu Breezy Badger (V5.01) box. Standard setup.
>
> - I downloaded brains2_RedHatEnterpriseWS4_20060110.030316.tar.gz, and
> unpacked to /opt/brains2. Ran pretty much out of the box.
>
> - All brains2 users are part of unix group 'office-users'. All of
> /opt/brains2 are readable by 'office-users'
>
> - Data is stored under an NFS mounted share at /data. Users of group
> 'office-users' have full write access.
>
> - Added the following to ~/.bashrc for each user: -
> --------------------------------------
> # Hacks for Brains2
> source /opt/brains2/bin/brains2_setup.sh
>
> # For some reasons, brains2 requires root access?
> alias brains2='sudo /opt/brains2/bin/brains2'
>
> # Change to default directory
> cd /data
>
> # Set Umask
> umask 002
> ----------------------------------------
>
> - Note that I had to run brains2 as 'root'!! (using sudo). The reason
> is that when Simone runs a test, at one stage it makes a directory and
> tries to 'chown' it, something only root can do. When the 'chown' fails,
> it appears to fail the script, and running it under root solved that.
> Not happy about that however.
>
> - Modified /opt/brains2/bin/brains2rc and changed the following: -
> -------------------------------------------
> B2_EVAL_TCL "B2_GLOBAL_DATA_DIR" "set B2_GLOBAL_DATA_DIR
> /data/new_data/brains2 ; "
> -------------------------------------------
>
>
> Cheers,
> Brendan
>
>
>
> -----Original Message-----
> From: Eugene Zeien [mailto:eugene-zeien at uiowa.edu]
> Sent: Thursday, 2 March 2006 11:31 PM
> To: Simone Brockman
> Subject: Re: [Brains-users] Error message during measurement of T50
> volumes
>
>
> Simone,
> Sorry. I told you how to do this using our latest version.
> With a brains2 rpm that is more than 2-3 months old, this line in
> brains/bin/brains2rc or ~/.brains2rc should redirect the measures
> to the specified directory "/home/morpho/brains2_data".
>
> B2_GLOBAL_DATA_DIR /home/morpho/brains2_data
>
> If you send a URL for a data set, I'll download it.
> I am uncertain whether I will have time to work/play with your MR and
> PET data. We are planning to have another "BRAINS camp" this year.
> The exact date has not been set, though.
>
>
> Simone Brockman wrote:
>
>>Thanks for that Eugene,
>>I have amended the default config file and the error still occurs..as it does when I change my .brains2rc file. I have spoken to our programmer who set up our system and he is thinking that it might have somthing to do with the way he set the system up initially... he's going to give you some info about exactly what he did, which I hope will shed some light.
>>
>>Some additional questions....we are starting to acquire PET and MRI's for one of our research projects
>>1. Can we send a copy of an MR image and a PET scan for you to check that all is okay ( format, quality etc). We are collaborating with Ricardo Jorge with one of our projects, and Sergio Starkstein (CI) thought to send the images to him, either via FTP or our website - can I use your FTP site to do this and would you (or someone in the lab) be looking at them anyway? or should I use our website and give the passwords asI did last time?..
>>2. I still would like you to take a look at one of our scans that couldn't be read here - again I'm just not sure of how to do that...I got a bit lost with your explanation when I last emailed you about this...sorry...I have a better awareness of the whole program and Linux system now, so hopefully i'll be able to follow your instructions..
>>3. Is there another BRAINS camp being run this year? - I will be in the US on holidays in August and would like to attend the course again if possible..
>>
>>Many thanks for your time
>>Simone
>>
>>
>>
>>
>>
>>-----Original Message-----
>>From: Eugene Zeien [mailto:eugene-zeien at uiowa.edu]
>>Sent: Wednesday, 22 February 2006 12:24 AM
>>To: Simone Brockman
>>Cc: brains-users at psychiatry.uiowa.edu
>>Subject: Re: [Brains-users] Error message during measurement of T50
>>volumes
>>
>>
>>In the default brains2 config file, /opt/brains2/bin/brains2rc
>>this line sets up where everyone dumps a copy of measures:
>>
>> B2_EVAL_TCL "B2_GLOBAL_DATA_DIR" "set B2_GLOBAL_DATA_DIR /raid0/data/new_data/brains2 ; "
>>
>>This line (if it weren't commented out) would configure where brains2
>>scripts are stored.
>>
>># Default Location to place batch files
>># B2_EVAL_TCL "B2_WORK_DIR" "set B2_WORK_DIR ${HOME}/<username> ; "
>>
>>If you wish to change these for everyone using your computer,
>>edit /opt/brains2/bin/brains2rc. If you want to change these settings
>>for yourself only,
>> copy these lines into your ~/.brains2rc
>> delete the # for the "B2_WORK_DIR" line
>> and change to locations of your choice
>>
>>For example,
>>
>> B2_EVAL_TCL "B2_GLOBAL_DATA_DIR" "set B2_GLOBAL_DATA_DIR /home/simone/brains2_data ; "
>> B2_EVAL_TCL "B2_WORK_DIR" "set B2_WORK_DIR /home/simone/scripts ; "
>>(might want to 'mkdir /volume/storage/location /home/simone/scripts' first) ;-)
>>
>>Simone Brockman wrote:
>>
>>
>>>It seems I spoke too soon about not having any further problems...
>>>
>>>When I go to run the t50 volumes, I get an error message both when I save it to run later and if I try to run it straight away. The error message when I try to save occurs after I try to save the _t50vol.brains2 in the default location of the home directory and reads
>>>"Failed to save the check measurement script! Verify there that there is a ...ACPC/root directory."
>>>
>>>and the error message when I try to run interactively occurs when I try to select the update button on the Check T50 Volumes dialogue box and reads
>>>
>>>"Failed to open global raw volume data file /raid0/data/new_data/brains2/t50/t50_volumes_raw.dat.
>>>
>>>I have the option to modify the Global Data File before running the T50 Volumes but it seems to be looking for something my system doesn't have. Isn't /raid0 an IPL specific directory structure? Am I missing a file or have I done soemthing incorrectly in preceding steps??
>>>
>>>Many thanks for your advice
>>>
>>>Simone
>>>UWA Perth
>>>Western Australia
>>>
>>>
>>>_______________________________________________
>>>Brains-users mailing list
>>>Brains-users at www.psychiatry.uiowa.edu
>>>http://www.psychiatry.uiowa.edu/mailman/listinfo/brains-users
>>>
>>
>>
>>
>>
>
>
>
More information about the Brains-users
mailing list