Compiling AFNI with cmake vs. make, missing mask.nii.gz file

npavlovikj · July 28, 2023, 3:51pm

Hi,

We compiled AFNI 23.1.10 with cmake on our Linux HPC system, and when running afni_proc.py with multiple test data I found, I keep getting error about mask.nii.gz not being generated:

++ requiring 5 voxels in mask columns
++ 3dLocalstat: AFNI version=AFNI_23.1.10 (Jul  8 2023) [64-bit]
++ Authored by: Emperor Zhark
** AFNI can't deal with 5 dimensional NIfTI(Flanker/sub-08.results/vlines.pb00.tcat/mask.nii.gz)
** FATAL ERROR: Can't open dataset 'mask.nii.gz'

To verify that this is not due to the installation, I also ran the same tests via Apptainer with the official AFNI Docker image you provide, docker://afni/afni_cmake_build:AFNI_23.1.10, and I encountered the same error about mask.nii.gz not being generated.

However, when I ran the same tests with the docker://afni/afni_make_build:AFNI_23.1.10 Docker image, the mask.nii.gz file was generated, and the tests finished successfully (although I got messages about bc: Command not found.).

Do you have any suggestions why using AFNI compiled with make works compared to AFNI compiled with cmake?
Is there an additional step I need to do with the cmake image/compilation in order to have it work correctly?
Or there is something else I am missing in my setup?

I am looking forward to hearing from you.
Please let me know if you need any additional information.

Thank you,
Natasha

ptaylor · July 28, 2023, 8:18pm

Hi, Natasha-

Hmm, I haven't seen those kinds of differences before. The make and cmake builds should match. It would help to see the output of afni_system_check.py -check_all for each build, if possible. That might point to a missing dependency in one case---I guess something in the cmake didn't build properly.

--pt

npavlovikj · July 28, 2023, 10:54pm

Hi @ptaylor ,

Thank you for your prompt reply!

Looks like I can not attach the requested files, so I will just paste the output below.

Output from afni_system_check.py when using local cmake built:

-------------------------------- general ---------------------------------
architecture:         64bit ELF
cpu type:             x86_64
system:               Linux
release:              4.18.0-477.13.1.el8_8.x86_64
version:              #1 SMP Tue May 30 14:53:41 EDT 2023
distribution:         AlmaLinux 8.8 (Sapphire Caracal)
number of CPUs:       1
apparent login shell: bash
shell RC file:        .bashrc (exists)

--------------------- AFNI and related program tests ---------------------
which afni           : /util/opt/anaconda/deployed-conda-envs/packages/afni/envs/afni-23.1.10/bin/afni
afni version         : Precompiled binary 1: Jul  8 2023 
                     : AFNI_23.1.10 'Publius Helvius Pertinax'
AFNI_version.txt     : AFNI_23.1.10, conda_centos7_64, Jul 08 2023, conda-build
which python         : /util/opt/anaconda/deployed-conda-envs/packages/afni/envs/afni-23.1.10/bin/python
python version       : 3.9.16
which R              : /util/opt/anaconda/deployed-conda-envs/packages/afni/envs/afni-23.1.10/bin/R
R version            : R version 4.2.3 (2023-03-15) -- "Shortstop Beagle"

instances of various programs found in PATH:
    afni    : 1   (/util/opt/anaconda/deployed-conda-envs/packages/afni/envs/afni-23.1.10/bin/afni)
    R       : 1   (/util/opt/anaconda/deployed-conda-envs/packages/afni/envs/afni-23.1.10/bin/R)
    python  : 1   (/util/opt/anaconda/deployed-conda-envs/packages/afni/envs/afni-23.1.10/bin/python3.9)
    python2 : 0 
    python3 : 2 
      /util/opt/anaconda/deployed-conda-envs/packages/afni/envs/afni-23.1.10/bin/python3.9
      /usr/libexec/platform-python3.6

** have python3 but not python2

testing ability to start various programs...
    afni                 : success
    suma                 : success
    3dSkullStrip         : success
    3dAllineate          : success
    3dRSFC               : success
    SurfMesh             : success
    3dClustSim           : success
    uber_subject.py      : success
    3dMVM                : success

------------------------ dependent program tests -------------------------
checking for dependent programs...

which tcsh           : /usr/bin/tcsh
which Xvfb           : /usr/bin/Xvfb

checking for R packages...
    rPkgsInstall -pkgs ALL -check : success

R RHOME : /util/opt/anaconda/deployed-conda-envs/packages/afni/envs/afni-23.1.10/lib/R

------------------------------ python libs -------------------------------

++ module loaded: matplotlib.pyplot
   module file : /util/opt/anaconda/deployed-conda-envs/packages/afni/envs/afni-23.1.10/lib/python3.9/site-packages/matplotlib/pyplot.py
   matplotlib version : 3.7.2

-------------------------------- env vars --------------------------------
PATH = /util/opt/anaconda/deployed-conda-envs/packages/afni/envs/afni-23.1.10/bin:/home/project/npavlovikj/.local/bin:/home/project/npavlovikj/bin:/usr/local/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/opt/puppetlabs/bin:/util/opt/bin

PYTHONPATH = 
R_LIBS = 
LD_LIBRARY_PATH = /util/opt/anaconda/deployed-conda-envs/packages/afni/envs/afni-23.1.10/x86_64-conda-linux-gnu/sysroot/usr/lib64

DYLD_LIBRARY_PATH = 
DYLD_FALLBACK_LIBRARY_PATH = 

----------------------------- eval dot files -----------------------------

----------- AFNI $HOME files -----------

    .afnirc                   : missing
    .sumarc                   : missing
    .afni/help/all_progs.COMP : missing

--------- shell startup files ----------

   -- good: .tcshrc seems to contain 'source .cshrc'
   -- considered operations: path, flatdir, apsearch
      (not on a mac, should skip flatdir)
   
   -- note: followers should not need edits, so edit flags should be 0
      (have 1 follower(s), which can be ignored)
   
   dot file test : want 2 modifications across 3 files:
   
      file             path  flatdir  apsearch        follower
      ---------------  ----  -------  --------        --------
      .cshrc           0     0        1               0     
      .tcshrc          0     0        0               1     
      .bashrc          0     0        1               0     
   
------------------------------ data checks -------------------------------
data dir : missing AFNI_data6
data dir : missing AFNI_demos
data dir : missing suma_demo
data dir : missing afni_handouts
atlas    : found TT_N27+tlrc  under /work/HCC/BCRF/app_specific/afni/afni_atlases_dist

------------------------------ OS specific -------------------------------
which dnf            : /usr/bin/dnf
dnf version          : 4.7.0
which yum            : /usr/bin/yum
yum version          : 4.7.0


=========================  summary, please fix:  =========================
*  just be aware: login shell 'bash', but our code examples use 'tcsh'
*  'afni' executable is owned by root
*  please run: cp /util/opt/anaconda/deployed-conda-envs/packages/afni/envs/afni-23.1.10/bin/AFNI.afnirc ~/.afnirc
*  please run: "suma -update_env" for .sumarc
*  please run: apsearch -update_all_afni_help
*  dot file test : want 2 modifications across 3 files:
*  insufficient data for AFNI bootcamp
   (see "Prepare for Bootcamp" on install pages)

Output from afni_system_check.py when using the Docker image built with cmake with Apptainer (apptainer shell docker://afni/afni_cmake_build:AFNI_23.1.10):

-------------------------------- general ---------------------------------
architecture:         64bit ELF
cpu type:             x86_64
system:               Linux
release:              4.18.0-477.13.1.el8_8.x86_64
version:              #1 SMP Tue May 30 14:53:41 EDT 2023
distribution:         Ubuntu 18.04 bionic
number of CPUs:       1
apparent login shell: bash
shell RC file:        .bashrc (exists)

--------------------- AFNI and related program tests ---------------------
which afni           :
                     : AFNI_99.99.99 'Alexios V Doukas the Bushy-eyebrowed'
AFNI_version.txt     : 
which python         :
which R              :

instances of various programs found in PATH:
    afni    : 1   (/opt/afni/install/usr/local/bin/afni)
    R       : 1   (/usr/bin/R)
    python  : 1   (/usr/bin/python3.6)
    python2 : 0 
    python3 : 1   (/usr/bin/python3.6)

** have python3 but not python2

testing ability to start various programs...
    afni                 : success
    suma                 : success
    3dSkullStrip         : success
    3dAllineate          : success
    3dRSFC               : success
    SurfMesh             : success
    3dClustSim           : success
    uber_subject.py      : success
    3dMVM                : FAILURE
        /bin/sh: 3dMVM: command not found

have failures, testing programs under implied /opt/user_pip_packages/bin...
    afni                 : FAILURE
        /bin/sh: /opt/user_pip_packages/bin/afni: No such file or directory
    suma                 : FAILURE
        /bin/sh: /opt/user_pip_packages/bin/suma: No such file or directory
    3dSkullStrip         : FAILURE
        /bin/sh: /opt/user_pip_packages/bin/3dSkullStrip: No such file or directory
    3dAllineate          : FAILURE
        /bin/sh: /opt/user_pip_packages/bin/3dAllineate: No such file or directory
    3dRSFC               : FAILURE
        /bin/sh: /opt/user_pip_packages/bin/3dRSFC: No such file or directory
    SurfMesh             : FAILURE
        /bin/sh: /opt/user_pip_packages/bin/SurfMesh: No such file or directory
    3dClustSim           : FAILURE
        /bin/sh: /opt/user_pip_packages/bin/3dClustSim: No such file or directory
    uber_subject.py      : success
    3dMVM                : FAILURE
        /bin/sh: /opt/user_pip_packages/bin/3dMVM: No such file or directory

------------------------ dependent program tests -------------------------
checking for dependent programs...

which tcsh           :
which Xvfb           :

checking for R packages...
    rPkgsInstall -pkgs ALL -check : FAILURE
        /bin/sh: rPkgsInstall: command not found

R RHOME : /usr/lib/R

------------------------------ python libs -------------------------------

++ module loaded: matplotlib.pyplot
   module file : /opt/user_pip_packages/lib/python3.6/site-packages/matplotlib/pyplot.py
   matplotlib version : 3.3.4

-------------------------------- env vars --------------------------------
PATH = /opt/afni/src/../install/usr/local/bin:/opt/cmake/cmake-3.14.7-Linux-x86_64/bin:/opt/user_pip_packages/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin

PYTHONPATH = 
R_LIBS = 
LD_LIBRARY_PATH = /.singularity.d/libs

DYLD_LIBRARY_PATH = 
DYLD_FALLBACK_LIBRARY_PATH = 

----------------------------- eval dot files -----------------------------

----------- AFNI $HOME files -----------

    .afnirc                   : missing
    .sumarc                   : missing
    .afni/help/all_progs.COMP : missing

--------- shell startup files ----------

   ** error: have no found abin, so please use -dir_bin
------------------------------ data checks -------------------------------
data dir : missing AFNI_data6
data dir : missing AFNI_demos
data dir : missing suma_demo
data dir : missing afni_handouts
atlas    : found TT_N27+tlrc  under /work/HCC/BCRF/app_specific/afni/afni_atlases_dist

------------------------------ OS specific -------------------------------

have Ubuntu system: Ubuntu 18.04 bionic

=========================  summary, please fix:  =========================
*  just be aware: login shell 'bash', but our code examples use 'tcsh'
*  missing program: afni
*  missing AFNI_version.txt, maybe package is old
*  missing program: python
*  missing program: R
*  failure under initial "AFNI and related program tests"
*  'afni' executable is owned by root
*  AFNI programs show FAILURE
*  consider adding /opt/user_pip_packages/bin to your PATH
*  missing program: tcsh
*  missing program: Xvfb
*  missing R packages (see rPkgsInstall)
*  please run: cp /opt/user_pip_packages/bin/AFNI.afnirc ~/.afnirc
*  please run: "suma -update_env" for .sumarc
*  please run: apsearch -update_all_afni_help
*  failure running init_user_dotfiles.py -test
*  insufficient data for AFNI bootcamp
   (see "Prepare for Bootcamp" on install pages)

Output from afni_system_check.py when using the Docker image built with make with Apptainer (apptainer shell docker://afni/afni_make_build:AFNI_23.1.10):

-------------------------------- general ---------------------------------
architecture:         64bit ELF
cpu type:             x86_64
system:               Linux
release:              4.18.0-477.13.1.el8_8.x86_64
version:              #1 SMP Tue May 30 14:53:41 EDT 2023
distribution:         Ubuntu 18.04 bionic
number of CPUs:       1
apparent login shell: bash
shell RC file:        .bashrc (exists)

--------------------- AFNI and related program tests ---------------------
which afni           :
                     : AFNI_23.1.10 'Publius Helvius Pertinax'
AFNI_version.txt     : AFNI_23.1.10, linux_ubuntu_16_64_glw_local_shared, Jun 30 2023, local
which python         :
which R              :

instances of various programs found in PATH:
    afni    : 1   (/opt/afni/install/afni)
    R       : 1   (/usr/bin/R)
    python  : 1   (/usr/bin/python3.6)
    python2 : 0 
    python3 : 1   (/usr/bin/python3.6)

** have python3 but not python2

testing ability to start various programs...
    afni                 : success
    suma                 : success
    3dSkullStrip         : success
    3dAllineate          : success
    3dRSFC               : success
    SurfMesh             : success
    3dClustSim           : success
    uber_subject.py      : success
    3dMVM                : success

------------------------ dependent program tests -------------------------
checking for dependent programs...

which tcsh           :
which Xvfb           :

checking for R packages...
    rPkgsInstall -pkgs ALL -check : FAILURE
        
        oo Warning: 
           These packages are not installed on the computer: afex!
        These packages are not installed on the computer: phia!
        These packages are not installed on the computer: snow!
        These packages are not installed on the computer: lmerTest!
        These packages are not installed on the computer: paran!
        These packages are not installed on the computer: psych!
        These packages are not installed on the computer: brms!
        These packages are not installed on the computer: corrplot!
        These packages are not installed on the computer: metafor!
        

R RHOME : /usr/lib/R

------------------------------ python libs -------------------------------

++ module loaded: matplotlib.pyplot
   module file : /opt/user_pip_packages/lib/python3.6/site-packages/matplotlib/pyplot.py
   matplotlib version : 3.3.4

-------------------------------- env vars --------------------------------
PATH = /opt/afni/src/../install:/opt/cmake/cmake-3.14.7-Linux-x86_64/bin:/opt/user_pip_packages/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin

PYTHONPATH = 
R_LIBS = 
LD_LIBRARY_PATH = /.singularity.d/libs

DYLD_LIBRARY_PATH = 
DYLD_FALLBACK_LIBRARY_PATH = 

----------------------------- eval dot files -----------------------------

----------- AFNI $HOME files -----------

    .afnirc                   : missing
    .sumarc                   : missing
    .afni/help/all_progs.COMP : missing

--------- shell startup files ----------

   ** error: have no found abin, so please use -dir_bin
------------------------------ data checks -------------------------------
data dir : missing AFNI_data6
data dir : missing AFNI_demos
data dir : missing suma_demo
data dir : missing afni_handouts
atlas    : found TT_N27+tlrc  under /work/HCC/BCRF/app_specific/afni/afni_atlases_dist

------------------------------ OS specific -------------------------------

have Ubuntu system: Ubuntu 18.04 bionic

=========================  summary, please fix:  =========================
*  just be aware: login shell 'bash', but our code examples use 'tcsh'
*  missing program: afni
*  missing program: python
*  missing program: R
*  failure under initial "AFNI and related program tests"
*  'afni' executable is owned by root
*  missing program: tcsh
*  missing program: Xvfb
*  missing R packages (see rPkgsInstall)
*  please run: cp /opt/afni/install/AFNI.afnirc ~/.afnirc
*  please run: "suma -update_env" for .sumarc
*  please run: apsearch -update_all_afni_help
*  failure running init_user_dotfiles.py -test
*  insufficient data for AFNI bootcamp
   (see "Prepare for Bootcamp" on install pages)

I am not AFNI user, so I was using a modified afni_proc.py command from the tutorial given here for 1 subject (sub-08) and I obtained all the necessary files following the earlier steps from the tutorial.

npavlovikj · July 28, 2023, 10:55pm

Please see below the output logs from tcsh -xef proc.sub-08 with the same data with the three different ways.

Log from tcsh -xef proc.sub-08 when using local cmake built:

echo auto-generated by afni_proc.py, Fri Jul 28 16:49:17 2023
auto-generated by afni_proc.py, Fri Jul 28 16:49:17 2023
echo (version 7.58, June 22, 2023)
(version 7.58, June 22, 2023)
echo execution started: `date`
date
execution started: Fri Jul 28 16:49:33 CDT 2023
afni -ver
Precompiled binary 1: Jul  8 2023 (Version AFNI_23.1.10 'Publius Helvius Pertinax')
afni_history -check_date 14 Nov 2022
-- is current: afni_history as new as: 14 Nov 2022
               most recent entry is:   26 Jun 2023
if ( 0 ) then
if ( 0 > 0 ) then
set subj = sub-08
endif
set output_dir = sub-08.results
if ( -d sub-08.results ) then
set runs = ( `count -digits 2 1 2` )
count -digits 2 1 2
mkdir -p sub-08.results
mkdir sub-08.results/stimuli
3dcopy /work/project/npavlovikj/afni/Flanker/sub-08/anat/sub-08_T1w.nii.gz sub-08.results/sub-08_T1w
++ 3dcopy: AFNI version=AFNI_23.1.10 (Jul  8 2023) [64-bit]
3dcopy /work/HCC/BCRF/app_specific/afni/afni_atlases_dist/MNI_avg152T1+tlrc.HEAD sub-08.results/MNI_avg152T1+tlrc.HEAD
++ 3dcopy: AFNI version=AFNI_23.1.10 (Jul  8 2023) [64-bit]
3dTcat -prefix sub-08.results/pb00.sub-08.r01.tcat /work/project/npavlovikj/afni/Flanker/sub-08/func/sub-08_task-flanker_run-1_bold.nii.gz[0..$]
++ 3dTcat: AFNI version=AFNI_23.1.10 (Jul  8 2023) [64-bit]
++ elapsed time = 0.7 s
3dTcat -prefix sub-08.results/pb00.sub-08.r02.tcat /work/project/npavlovikj/afni/Flanker/sub-08/func/sub-08_task-flanker_run-2_bold.nii.gz[0..$]
++ 3dTcat: AFNI version=AFNI_23.1.10 (Jul  8 2023) [64-bit]
++ elapsed time = 0.7 s
set tr_counts = ( 146 146 )
cd sub-08.results
tee out.vlines.pb00.tcat.txt
find_variance_lines.tcsh -polort 2 -nerode 2 -rdir vlines.pb00.tcat pb00.sub-08.r01.tcat+orig.HEAD pb00.sub-08.r02.tcat+orig.HEAD
++ have nslices : 40 40
++ copying pb00.sub-08.r01.tcat+orig.HEAD[0..$] to ts.0.orig.r01.nii.gz
++ 3dTcat: AFNI version=AFNI_23.1.10 (Jul  8 2023) [64-bit]
++ elapsed time = 18.4 s
++ copying pb00.sub-08.r02.tcat+orig.HEAD[0..$] to ts.0.orig.r02.nii.gz
++ 3dTcat: AFNI version=AFNI_23.1.10 (Jul  8 2023) [64-bit]
++ elapsed time = 11.9 s
++ creating automask from first dset
++ 3dAutomask: AFNI version=AFNI_23.1.10 (Jul  8 2023) [64-bit]
++ Authored by: Emperor Zhark
++ Loading dataset ts.0.orig.r01.nii.gz
++ Forming automask
 + Fixed clip level = 343.908478
 + Used gradual clip level = 328.785889 .. 363.965637
 + Number voxels above clip level = 39446
 + Clustering voxels ...
 + Largest cluster has 38882 voxels
 + Clustering voxels ...
 + Largest cluster has 38556 voxels
 + Filled   356 voxels in small holes; now have 38912 voxels
 + Filled     3 voxels in large holes; now have 38915 voxels
 + Clustering voxels ...
 + Largest cluster has 38915 voxels
 + Clustering non-brain voxels ...
 + Clustering voxels ...
 + Largest cluster has 124925 voxels
 + Mask now has 38915 voxels
++ 38915 voxels in the mask [out of 163840: 23.75%]
++ first  11 x-planes are zero [from L]
++ last   11 x-planes are zero [from R]
++ first   0 y-planes are zero [from P]
++ last    5 y-planes are zero [from A]
++ first   1 z-planes are zero [from I]
++ last    3 z-planes are zero [from S]
++ Output dataset ./mask.nii.gz
++ CPU time = 0.000000 sec

-- will apply mask mask.nii.gz
++ eroding mask.nii.gz by 2 voxels
++ no -frac option: defaulting to -union
++ processing 1 input dataset(s), NN=2...
++ padding all datasets by 0 (for dilations)
++ frac 0 over 1 volumes gives min count 0
++ voxel limits: 0 clipped, 23179 survived, 140661 were zero
++ writing result mask.nii.gz...
++ Output dataset ./mask.nii.gz

++ requiring 5 voxels in mask columns
++ 3dLocalstat: AFNI version=AFNI_23.1.10 (Jul  8 2023) [64-bit]
++ Authored by: Emperor Zhark
** AFNI can't deal with 5 dimensional NIfTI(/lustre/work/project/npavlovikj/afni/Flanker/sub-08.results/vlines.pb00.tcat/mask.nii.gz)
** FATAL ERROR: Can't open dataset 'mask.nii.gz'
** Program compile date = Jul  8 2023
++ 3dcalc: AFNI version=AFNI_23.1.10 (Jul  8 2023) [64-bit]
++ Authored by: A cast of thousands
** FATAL ERROR: can't open dataset tmp.mask.col.count.nii.gz
** Program compile date = Jul  8 2023

-- detrend -polort 2, new eset = ts.1.det.r01.nii.gz
++ 3dTproject: AFNI version=AFNI_23.1.10 (Jul  8 2023) [64-bit]
++ Authored by: Cox the Algebraic (Linear)
++ Setting up regressors
++ 1 Blocks * 3 polynomials -- 3 polort regressors
++ 146 retained time points MINUS 3 regressors ==> 143 D.O.F. left
++ no -mask option ==> processing all 163840 voxels in dataset
*+ WARNING: 3dTproject input data :: 2561 vectors are constant

++ 3dTstat: AFNI version=AFNI_23.1.10 (Jul  8 2023) [64-bit]
++ Authored by: KR Hammett & RW Cox
++ Output dataset ./tmp.stdev.nii.gz
++ 3dcalc: AFNI version=AFNI_23.1.10 (Jul  8 2023) [64-bit]
++ Authored by: A cast of thousands
++ Output dataset ./var.0.orig.r01.nii.gz
** FATAL ERROR: Can't open dataset 'mask.nii.gz'
** Program compile date = Jul  8 2023

Log from tcsh -xef proc.sub-08 when using the Docker image built with cmake with Apptainer (apptainer shell docker://afni/afni_cmake_build:AFNI_23.1.10):

echo auto-generated by afni_proc.py, Fri Jul 28 16:56:00 2023
auto-generated by afni_proc.py, Fri Jul 28 16:56:00 2023
echo (version 7.58, June 22, 2023)
(version 7.58, June 22, 2023)
echo execution started: `date`
date
execution started: Fri Jul 28 16:56:10 CDT 2023
afni -ver
Precompiled binary 1: Jun 30 2023 (Version AFNI_99.99.99 'Alexios V Doukas the Bushy-eyebrowed')
afni_history -check_date 14 Nov 2022
-- is current: afni_history as new as: 14 Nov 2022
               most recent entry is:   26 Jun 2023
if ( 0 ) then
if ( 0 > 0 ) then
set subj = sub-08
endif
set output_dir = sub-08.results
if ( -d sub-08.results ) then
set runs = ( `count -digits 2 1 2` )
count -digits 2 1 2
mkdir -p sub-08.results
mkdir sub-08.results/stimuli
3dcopy /work/project/npavlovikj/afni/Flanker/sub-08/anat/sub-08_T1w.nii.gz sub-08.results/sub-08_T1w
++ 3dcopy: AFNI version=AFNI_99.99.99 (Jun 30 2023) [64-bit]
3dcopy /work/HCC/BCRF/app_specific/afni/afni_atlases_dist/MNI_avg152T1+tlrc.HEAD sub-08.results/MNI_avg152T1+tlrc.HEAD
++ 3dcopy: AFNI version=AFNI_99.99.99 (Jun 30 2023) [64-bit]
3dTcat -prefix sub-08.results/pb00.sub-08.r01.tcat /work/project/npavlovikj/afni/Flanker/sub-08/func/sub-08_task-flanker_run-1_bold.nii.gz[0..$]
++ 3dTcat: AFNI version=AFNI_99.99.99 (Jun 30 2023) [64-bit]
++ elapsed time = 0.7 s
3dTcat -prefix sub-08.results/pb00.sub-08.r02.tcat /work/project/npavlovikj/afni/Flanker/sub-08/func/sub-08_task-flanker_run-2_bold.nii.gz[0..$]
++ 3dTcat: AFNI version=AFNI_99.99.99 (Jun 30 2023) [64-bit]
++ elapsed time = 0.7 s
set tr_counts = ( 146 146 )
cd sub-08.results
tee out.vlines.pb00.tcat.txt
find_variance_lines.tcsh -polort 2 -nerode 2 -rdir vlines.pb00.tcat pb00.sub-08.r01.tcat+orig.HEAD pb00.sub-08.r02.tcat+orig.HEAD
++ have nslices : 40 40
++ copying pb00.sub-08.r01.tcat+orig.HEAD[0..$] to ts.0.orig.r01.nii.gz
++ 3dTcat: AFNI version=AFNI_99.99.99 (Jun 30 2023) [64-bit]
++ elapsed time = 4.3 s
++ copying pb00.sub-08.r02.tcat+orig.HEAD[0..$] to ts.0.orig.r02.nii.gz
++ 3dTcat: AFNI version=AFNI_99.99.99 (Jun 30 2023) [64-bit]
++ elapsed time = 4.2 s
++ creating automask from first dset
++ 3dAutomask: AFNI version=AFNI_99.99.99 (Jun 30 2023) [64-bit]
++ Authored by: Emperor Zhark
++ Loading dataset ts.0.orig.r01.nii.gz
++ Forming automask
 + Fixed clip level = 343.908478
 + Used gradual clip level = 328.785889 .. 363.965637
 + Number voxels above clip level = 39446
 + Clustering voxels ...
 + Largest cluster has 38882 voxels
 + Clustering voxels ...
 + Largest cluster has 38556 voxels
 + Filled   356 voxels in small holes; now have 38912 voxels
 + Filled     3 voxels in large holes; now have 38915 voxels
 + Clustering voxels ...
 + Largest cluster has 38915 voxels
 + Clustering non-brain voxels ...
 + Clustering voxels ...
 + Largest cluster has 124925 voxels
 + Mask now has 38915 voxels
++ 38915 voxels in the mask [out of 163840: 23.75%]
++ first  11 x-planes are zero [from L]
++ last   11 x-planes are zero [from R]
++ first   0 y-planes are zero [from P]
++ last    5 y-planes are zero [from A]
++ first   1 z-planes are zero [from I]
++ last    3 z-planes are zero [from S]
++ Output dataset ./mask.nii.gz
++ CPU time = 0.000000 sec

-- will apply mask mask.nii.gz
++ eroding mask.nii.gz by 2 voxels
++ no -frac option: defaulting to -union
++ processing 1 input dataset(s), NN=2...
++ padding all datasets by 0 (for dilations)
++ frac 0 over 1 volumes gives min count 0
++ voxel limits: 0 clipped, 23179 survived, 140661 were zero
++ writing result mask.nii.gz...
++ Output dataset ./mask.nii.gz

++ requiring 5 voxels in mask columns
++ 3dLocalstat: AFNI version=AFNI_99.99.99 (Jun 30 2023) [64-bit]
++ Authored by: Emperor Zhark
** AFNI can't deal with 5 dimensional NIfTI(/work/project/npavlovikj/afni/Flanker/sub-08.results/vlines.pb00.tcat/mask.nii.gz)
** FATAL ERROR: Can't open dataset 'mask.nii.gz'
** Program compile date = Jun 30 2023
++ 3dcalc: AFNI version=AFNI_99.99.99 (Jun 30 2023) [64-bit]
++ Authored by: A cast of thousands
** FATAL ERROR: can't open dataset tmp.mask.col.count.nii.gz
** Program compile date = Jun 30 2023

-- detrend -polort 2, new eset = ts.1.det.r01.nii.gz
++ 3dTproject: AFNI version=AFNI_99.99.99 (Jun 30 2023) [64-bit]
++ Authored by: Cox the Algebraic (Linear)
++ Setting up regressors
++ 1 Blocks * 3 polynomials -- 3 polort regressors
++ 146 retained time points MINUS 3 regressors ==> 143 D.O.F. left
++ no -mask option ==> processing all 163840 voxels in dataset
*+ WARNING: 3dTproject input data :: 2561 vectors are constant

++ 3dTstat: AFNI version=AFNI_99.99.99 (Jun 30 2023) [64-bit]
++ Authored by: KR Hammett & RW Cox
++ Output dataset ./tmp.stdev.nii.gz
++ 3dcalc: AFNI version=AFNI_99.99.99 (Jun 30 2023) [64-bit]
++ Authored by: A cast of thousands
++ Output dataset ./var.0.orig.r01.nii.gz
** FATAL ERROR: Can't open dataset 'mask.nii.gz'
** Program compile date = Jun 30 2023

npavlovikj · July 28, 2023, 11:01pm

Log from tcsh -xef proc.sub-08 when using the Docker image built with make with Apptainer (apptainer shell docker://afni/afni_make_build:AFNI_23.1.10) (this is a very long output, so I am just pasting the beginning and the end of it so you see that it differs from the previous two):

echo auto-generated by afni_proc.py, Fri Jul 28 17:09:02 2023
auto-generated by afni_proc.py, Fri Jul 28 17:09:02 2023
echo (version 7.58, June 22, 2023)
(version 7.58, June 22, 2023)
echo execution started: `date`
date
execution started: Fri Jul 28 17:09:10 CDT 2023
afni -ver
Precompiled binary linux_ubuntu_16_64_glw_local_shared: Jun 30 2023 (Version AFNI_23.1.10 'Publius Helvius Pertinax')
afni_history -check_date 14 Nov 2022
-- is current: afni_history as new as: 14 Nov 2022
               most recent entry is:   26 Jun 2023
if ( 0 ) then
if ( 0 > 0 ) then
set subj = sub-08
endif
set output_dir = sub-08.results
if ( -d sub-08.results ) then
set runs = ( `count -digits 2 1 2` )
count -digits 2 1 2
mkdir -p sub-08.results
mkdir sub-08.results/stimuli
3dcopy /work/project/npavlovikj/afni/Flanker/sub-08/anat/sub-08_T1w.nii.gz sub-08.results/sub-08_T1w
++ 3dcopy: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
3dcopy /work/HCC/BCRF/app_specific/afni/afni_atlases_dist/MNI_avg152T1+tlrc.HEAD sub-08.results/MNI_avg152T1+tlrc.HEAD
++ 3dcopy: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
3dTcat -prefix sub-08.results/pb00.sub-08.r01.tcat /work/project/npavlovikj/afni/Flanker/sub-08/func/sub-08_task-flanker_run-1_bold.nii.gz[0..$]
++ 3dTcat: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ elapsed time = 0.7 s
3dTcat -prefix sub-08.results/pb00.sub-08.r02.tcat /work/project/npavlovikj/afni/Flanker/sub-08/func/sub-08_task-flanker_run-2_bold.nii.gz[0..$]
++ 3dTcat: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ elapsed time = 0.7 s
set tr_counts = ( 146 146 )
cd sub-08.results
tee out.vlines.pb00.tcat.txt
find_variance_lines.tcsh -polort 2 -nerode 2 -rdir vlines.pb00.tcat pb00.sub-08.r01.tcat+orig.HEAD pb00.sub-08.r02.tcat+orig.HEAD
++ have nslices : 40 40
++ copying pb00.sub-08.r01.tcat+orig.HEAD[0..$] to ts.0.orig.r01.nii.gz
++ 3dTcat: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ elapsed time = 5.7 s
++ copying pb00.sub-08.r02.tcat+orig.HEAD[0..$] to ts.0.orig.r02.nii.gz
++ 3dTcat: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ elapsed time = 4.4 s
++ creating automask from first dset
++ 3dAutomask: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Authored by: Emperor Zhark
++ Loading dataset ts.0.orig.r01.nii.gz
++ Forming automask
 + Fixed clip level = 343.908478
 + Used gradual clip level = 328.785889 .. 363.965637
 + Number voxels above clip level = 39446
 + Clustering voxels ...
 + Largest cluster has 38882 voxels
 + Clustering voxels ...
 + Largest cluster has 38556 voxels
 + Filled   356 voxels in small holes; now have 38912 voxels
 + Filled     3 voxels in large holes; now have 38915 voxels
 + Clustering voxels ...
 + Largest cluster has 38915 voxels
 + Clustering non-brain voxels ...
 + Clustering voxels ...
 + Largest cluster has 124925 voxels
 + Mask now has 38915 voxels
++ 38915 voxels in the mask [out of 163840: 23.75%]
++ first  11 x-planes are zero [from L]
++ last   11 x-planes are zero [from R]
++ first   0 y-planes are zero [from P]
++ last    5 y-planes are zero [from A]
++ first   1 z-planes are zero [from I]
++ last    3 z-planes are zero [from S]
++ Output dataset ./mask.nii.gz
++ CPU time = 0.000000 sec

-- will apply mask mask.nii.gz
++ eroding mask.nii.gz by 2 voxels
++ no -frac option: defaulting to -union
++ processing 1 input dataset(s), NN=2...
++ padding all datasets by 0 (for dilations)
++ frac 0 over 1 volumes gives min count 0
++ voxel limits: 0 clipped, 23179 survived, 140661 were zero
++ writing result mask.nii.gz...
++ Output dataset ./mask.nii.gz

++ requiring 5 voxels in mask columns
++ 3dLocalstat: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Authored by: Emperor Zhark
++ Number of voxels in mask = 23179
++ Neighborhood comprises 81 voxels
++ Start sub-brick [0]
++ Output dataset ./tmp.mask.col.count.nii.gz
++ 3dcalc: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Authored by: A cast of thousands
++ Output dataset ./mask.nii.gz

-- detrend -polort 2, new eset = ts.1.det.r01.nii.gz
++ 3dTproject: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Authored by: Cox the Algebraic (Linear)
++ Setting up regressors
++ 1 Blocks * 3 polynomials -- 3 polort regressors
++ 146 retained time points MINUS 3 regressors ==> 143 D.O.F. left
++ no -mask option ==> processing all 163840 voxels in dataset
*+ WARNING: 3dTproject input data :: 2561 vectors are constant

++ 3dTstat: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Authored by: KR Hammett & RW Cox
++ Output dataset ./tmp.stdev.nii.gz
++ 3dcalc: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Authored by: A cast of thousands
++ Output dataset ./var.0.orig.r01.nii.gz
++ 3dcalc: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Authored by: A cast of thousands
++ Output dataset ./var.1.scale.r01.nii.gz
++ 3dLocalstat: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Authored by: Emperor Zhark
++ Number of voxels in mask = 22866
++ Neighborhood comprises 81 voxels
++ Start sub-brick [0]
++ Output dataset ./proj.r01.nii.gz
++ Neighborhood definition (NN=3) accepted
++ User input stat threshold: -1
++ User input stat threshold: 0.95
++ Data volume:      [0] 'MEAN'
++ Threshold volume: [0] 'MEAN'
++ How many sides to this stat? 0
++ Looks like clustering is *not* being performed on a stat value; won't worry about sidedness, then.
++ Opt code: 2sided,-1,0.95,NN3
*+ WARNING: No mask being used? That *could* be OK, but thought I'd let you know...
++ Threshold volume [0] does *not* appear to be a stat!
++ Sorting clusters by size.
#
#  Cluster report 
#[ Dataset prefix      = proj.r01.nii.gz ]
#[ Threshold vol       = [0] 'MEAN' ]
#[ Supplement dat vol  = [0] 'MEAN' ]
#[ Option summary      = 2sided,-1,0.95,NN3 ]
#[ Threshold value(s)  = left-tail thr=-1.000000;  right-tail thr=0.950000 ]
#[ Aux. stat. info.    = not a stat! ]
#[ Nvoxel threshold    = 0;  Volume threshold = 0.000 ]
#[ Single voxel volume = 36.000 (microliters) ]
#[ Neighbor type, NN   = 3 ]
#[ Voxel datum type    = float ]
#[ Voxel dimensions    = 3.000 mm X 3.000 mm X 4.000 mm ]
#[ Coordinates Order   = RAI ]
#[ Mean and SEM based on signed voxel intensities ]
#
#Volume  CM RL  CM AP  CM IS  minRL  maxRL  minAP  maxAP  minIS  maxIS    Mean     SEM    Max Int  MI RL  MI AP  MI IS
#------  -----  -----  -----  -----  -----  -----  -----  -----  -----  -------  -------  -------  -----  -----  -----
     10   -4.5  -36.1    3.8   -4.5   -4.5  -36.1  -36.1  -14.2   21.8   0.9663        0   0.9663   -4.5  -36.1  -14.2 
  -4.50  -36.10  -20.24
-- detrend -polort 2, new eset = ts.1.det.r02.nii.gz
++ 3dTproject: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Authored by: Cox the Algebraic (Linear)
++ Setting up regressors
++ 1 Blocks * 3 polynomials -- 3 polort regressors
++ 146 retained time points MINUS 3 regressors ==> 143 D.O.F. left
++ no -mask option ==> processing all 163840 voxels in dataset
*+ WARNING: 3dTproject input data :: 2560 vectors are constant

++ 3dTstat: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Authored by: KR Hammett & RW Cox
++ Output dataset ./tmp.stdev.nii.gz
++ 3dcalc: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Authored by: A cast of thousands
++ Output dataset ./var.0.orig.r02.nii.gz
++ 3dcalc: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Authored by: A cast of thousands
++ Output dataset ./var.1.scale.r02.nii.gz
++ 3dLocalstat: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Authored by: Emperor Zhark
++ Number of voxels in mask = 22866
++ Neighborhood comprises 81 voxels
++ Start sub-brick [0]
++ Output dataset ./proj.r02.nii.gz
++ Neighborhood definition (NN=3) accepted
++ User input stat threshold: -1
++ User input stat threshold: 0.95
++ Data volume:      [0] 'MEAN'
++ Threshold volume: [0] 'MEAN'
++ How many sides to this stat? 0
++ Looks like clustering is *not* being performed on a stat value; won't worry about sidedness, then.
++ Opt code: 2sided,-1,0.95,NN3
*+ WARNING: No mask being used? That *could* be OK, but thought I'd let you know...
++ Threshold volume [0] does *not* appear to be a stat!
++ Sorting clusters by size.
#
...
------------------ start of optionizing ------------------

++ Found input file:   ./MNI_avg152T1+tlrc.HEAD
++ Found input file:   TSNR.sub-08.rh.niml.dset
++ Found focus refbox file:   ./MNI_avg152T1+tlrc.HEAD
++ Using blowup factor: 2
++ Using opacity:  5
++ Making temporary work directory to copy vis files: QC_sub-08/media/__tmp_chauf_TryUDLvDHHv
++ Converted 0 to labels
++ Final subbrick indices: 0 0 0
bc: Command not found.
++ Copy ulay to visualize (volumetric) within user's range:
bc: Command not found.
bc: Command not found.
bc: Command not found.
++ Ulay to be visualized within user range:
  [0%, 120%] -> [0.000000, ]
++ 3dcalc: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Authored by: A cast of thousands
++ Output dataset ./QC_sub-08/media/__tmp_chauf_TryUDLvDHHv/tmp_ulay.nii
++ 3dcalc: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Authored by: A cast of thousands
++ Output dataset ./QC_sub-08/media/__tmp_chauf_TryUDLvDHHv/tmp_olay.nii
++ User-entered function range value value ()
++ Dimensions (xyzt): 91 109 91 1
++ (initial) Slice spacing ordered (x,y,z) is:  13 15 13
++ 3dAutobox: AFNI version=AFNI_23.1.10 (Jun 30 2023) [64-bit]
++ Auto bbox: x=10..81  y=10..101  z=3..78
++ 3dAutobox: output dataset = QC_sub-08/media/__tmp_chauf_TryUDLvDHHv/ulay_box_0.nii
bc: Command not found.
bc: Command not found.
bc: Command not found.
bc: Command not found.
@: Expression Syntax.
++ How many coors? 0
++ Will have the ref box central coors : SET_DICOM_XYZ 
++ Will have the ref box central gapord: 10 13 10

------------------- end of optionizing -------------------

 -- trying to start Xvfb :599 
[1] 3760717
_XSERVTransmkdir: Owner of /tmp/.X11-unix should be set to root
** nifti_header_version: bad sizeof_hdr = 1094459427
** nifti_image_read: bad nifti im header version -1
+/work/project/npavlovikj/afni/Flanker/sub-08.results/QC_sub-08/media/__tmp_chauf_TryUDLvDHHv*+ WARNING: Bad drive AFNI result from 'SWITCH_OVERLAY  tmp_olay.nii'
++ Writing palette image to QC_sub-08/media/qc_08_regr_tsnr_fin.pbar.jpg
*+ WARNING: Bad drive AFNI result from 'SET_FUNC_RANGE'
*+ WARNING: Bad drive AFNI result from 'SET_DICOM_XYZ'
++ Writing one 1286x218 image to filter '/opt/afni/src/../install/cjpeg -quality 95 > QC_sub-08/media/qc_08_regr_tsnr_fin.axi.jpg'
++ Writing one 1538x182 image to filter '/opt/afni/src/../install/cjpeg -quality 95 > QC_sub-08/media/qc_08_regr_tsnr_fin.sag.jpg'

AFNI QUITTs!

+++ Command Echo:
   afni -q -no1D -noplugins -no_detach -com SWITCH_UNDERLAY tmp_ulay.nii -com SWITCH_OVERLAY  tmp_olay.nii -com SEE_OVERLAY     + -com OPEN_WINDOW sagittalimage opacity=5 mont=7x1:10:1:black crop=0:0,0:0  -com OPEN_WINDOW coronalimage  opacity=5 mont=7x1:13:1:black crop=0:0,0:0  -com OPEN_WINDOW axialimage    opacity=5 mont=7x1:10:1:black crop=0:0,0:0  -com SET_PBAR_ALL    +99 1 Viridis -com PBAR_SAVEIM QC_sub-08/media/qc_08_regr_tsnr_fin.pbar.jpg dim=64x512H -com SET_SUBBRICKS   0 0 0 -com SET_ULAY_RANGE A.all 0.000000  -com SET_FUNC_RANGE   -com SET_THRESHNEW   0 * -com SET_FUNC_ALPHA  No -com SET_FUNC_BOXED  No -com SET_FUNC_RESAM  NN.NN -com SET_XHAIRS      OFF -com SET_XHAIR_GAP   -1 -com SET_DICOM_XYZ  -com SAVE_JPEG axialimage    QC_sub-08/media/qc_08_regr_tsnr_fin.axi blowup=2 -com SAVE_JPEG sagittalimage QC_sub-08/media/qc_08_regr_tsnr_fin.sag blowup=2 -com QUITT QC_sub-08/media/__tmp_chauf_TryUDLvDHHv 


+* Removing temporary image directory 'QC_sub-08/media/__tmp_chauf_TryUDLvDHHv'.

[1]  + Done                          Xvfb :599 -screen 0 1024x768x24

++ DONE (good exit)
   see: QC_sub-08/media/qc_08_regr_tsnr_fin*

++ Check for corr matrix warnings in: QC_sub-08/media/qc_09_warns_xmat.dat
++ Check EPI variance lines (vlines) in: vlines.pb00.tcat
++ Check basic summary quants from proc in: QC_sub-08/media/qc_12_qsumm_ssrev.dat


# +++++++++++ Check output of @ss_review_basic +++++++++++ #


subject ID                : sub-08
AFNI version              : AFNI_23.1.10
AFNI package              : linux_ubuntu_16_64_glw_local_shared
TR                        : 2.0
TRs removed (per run)     : 0
multiband level           : 1
slice timing pattern      : simult
num stim classes provided : 0
final anatomy dset        : anat_final.sub-08+tlrc.HEAD
final stats dset          : NO_STATS
final errts dset          : errts.sub-08.rh.tproject.niml.dset
orig voxel counts         : 64	64	40
orig voxel resolution     : 3.000000	3.000000	4.000000
orig volume center        : 0.000000	1.445793	-20.240967
final voxel resolution    : 1.000000	1.000000	1.000000

motion limit              : 0.3
num TRs above mot limit   : 2
average motion (per TR)   : 0.076751
max motion displacement   : 0.868833
average outlier frac (TR) : 0.00122644

num runs found            : 2
num TRs per run           : 146 146
TRs total                 : 292
degrees of freedom used   : 12
degrees of freedom left   : 280
final DF fraction         : 0.958904

num regs of interest      : 0

TSNR average              : 98.9689

apqc_make_html.py -qc_dir QC_sub-08

++ Done! Wrote QC HTML.  To check, consider:

   afni_open -b /work/project/npavlovikj/afni/Flanker/sub-08.results/QC_sub-08/index.html

echo \nconsider running: \n

consider running: 

echo    afni_open -b sub-08.results/QC_sub-08/index.html
   afni_open -b sub-08.results/QC_sub-08/index.html
echo 

endif
cd ..
echo execution finished: `date`
date
execution finished: Fri Jul 28 17:19:48 CDT 2023

npavlovikj · July 28, 2023, 11:03pm

I am sorry for the many replies - I exceeded the character limit when trying to post all the outputs in one message.

I hope this helps, and please let me know if you need any additional information.

Thank you,
Natasha

ptaylor · July 30, 2023, 2:26pm

Thanks for sending that.

The only previous time I have seen that specific error message:

** AFNI can't deal with 5 dimensional NIfTI(...)
** FATAL ERROR: Can't open dataset 'mask.nii.gz'

was when a user on an HPC reported it, but unfortunately we didn't hear back further information in the discussion.

I am not very familiar with Docker building, and haven't tried Apptainer. #1 and #2 above each have failures early on, and though #3 gets by that point, the fact that bc, tcsh and python are all missing (as well as 'afni') in it, is troublesome.

My first question would be, is using a precompiled binary not an issue, or running a local build without a container?

Assuming that containerization is necessary, is using Apptainer a requirement? We do have some docker notes for containerizing, I'm not sure if those might provide an alternative to using Apptainer?
https://afni.nimh.nih.gov/pub/dist/doc/htmldoc/background_install/container.html
That is what we use for our CircleCI (continuous integration) testing on GitHub.

But if containinerizing with Apptainer is specifically required, we can look more at hte current system.

--pt

npavlovikj · August 1, 2023, 10:59pm

Hi,

The results from #1 are from a local build using cmake (packaged via conda), while #2 and #3 are using the Docker images you officially provide with Apptainer (former Singularity).

Using Docker requires root permissions, which we don't give to users on our HPC system. On the other hand, Apptainer (former Singularity) is a containerized solution designed for HPC environments. It allows running images from variety of sources, including Docker images, and it doesn't require root permissions. Our researchers are successfully running various Docker images on our HPC system using Apptainer (such as neurodocker and freesurfer).

To test that Apptainer is not the reason for the error, I ran the two Docker images of AFNI 23.1.10 on my Fedora Virtual Machine using only Docker following the instructions you shared with me. I encountered the same exact errors with using Docker as I did when using Docker with Apptainer.

$ docker run -it --rm -v ${PWD}:/opt/home  afni/afni_cmake_build:AFNI_23.1.10

https:// gist. github. com/ npavlovikj/ b81735275e9c8923e8d216fa8d7881a2

$ docker run -it --rm -v ${PWD}:/opt/home  afni/afni_make_build:AFNI_23.1.10

https:// gist. github. com/ npavlovikj/ 8a823828486c4c19eb5c5f7be342521d

Are you able to replicate the errors I am encountering when you use the two Docker images, afni/afni_cmake_build:AFNI_23.1.10 and afni/afni_make_build:AFNI_23.1.10, on your test dataset yourself?

I further tested the two Docker images for AFNI 23.1.08 and got the same errors.

On the other hand, using the two Docker images for AFNI 22.2.12 worked (at least there was no error about mask.nii.gz not found which is the main error I am encountering).

Based on all of this, looks like it may be only the AFNI 23.1 cmake and make builds that are showing different results.

npavlovikj · August 1, 2023, 11:01pm

We did look for differences in the files in both Docker images with docker run -it --rm afni/afni_cmake_build:AFNI_23.1.10 find /opt/afni/install > cmake_out.txt and docker run -it --rm afni/afni_make_build:AFNI_23.1.10 find /opt/afni/install > make_out.txt, and the file with these differences can be found at https: //gist. github. com/ npavlovikj/ 33b07417966e779d75691bbf688fbc3a.
As you can see, there are many files that differ between the two Docker builds for AFNI 23.1.10.

I saw there was a new recent release of AFNI, 23.2.02. There are no official Docker images available for this release yet, so I couldn't test them. However, we packaged the binaries in a Docker image and used it with Apptainer on our HPC system, and that looks like it worked with the test data we have (at least there was no error about mask.nii.gz not found which as I said is the main error I am encountering).

I hope this helps.
If you need any further information, or there are some other things you want me to test, please let me know.

Thank you,
Natasha

ptaylor · June 15, 2024, 2:26pm

Hi, Natasha-

Just to let you know, we have (at long last) updated our CircleCI infrastructure and automatic Docker building has started again. There is a new version available (ttps://hub.docker.com/u/afni) from a build yesterday, if that is of interest. This was due to the heroic efforts of @yarikoptic and @rickr.

We have also started creating a new Docker-build script from scratch. We want to keep this process as simple as possible, to avoid long lapses; and we also plan to include R in one version of it. Hopefully this will also be useful in your projects.

--pt