Troubleshooting and Known Issues

1. yt compilation complains with something like: “Error compiling Cython file”

Answer - try running:

pip install -U Cython

2. Python-fsps compilation returns errors along the lines of

Error:

compiling Fortran sources
Fortran f77 compiler: /usr/bin/gfortran -Wall -ffixed-form -fno-second-underscore -fPIC -O3 -funroll-loops
Fortran f90 compiler: /usr/bin/gfortran -fPIC -fPIC -O3 -funroll-loops

Try re-compiling with the flag:

-fPIC

2. Python-fsps compilation can’t find ‘f95’

Error (see here):

Could not locate executable f95
...
error: Command "f95 -fPIC -fPIC -O3 ..." failed with exit status 127

Try setting an alias to gfortran in your .bashrc:

alias f95='gfortran'

If this doesn’t work, create a soft link to the gfortran executable in your path, e.g.:

ln -s /apps/compilers/gcc/6.3.0/bin/gfortran ~/bin/f95

3. When running pd via a SLURM scheduler, you get the error when importing fsps

Error:

build/bdist.linux-x86_64/egg/fsps/__init__.py in <module>()
ImportError: Your FSPS version does not seem to be under git version control. FSPS is available on github at https://github.com/cconroy20/fsps and should be cloned from there

Comment out the lines in python-fsps/fsps/__init__.py surrounding the checking of githashes. (h/t to Ena Choi for uncovering this one).

4. If in powderday you start getting memory errors on the Pool process call that look like this:

Error:

calculating the SEDs for  42  bins
Traceback (most recent call last):
File "/ufrc/narayanan/desika.narayanan/pd/pd_front_end.py", line 139, in <module>
  m=add_binned_seds(df_nu,stars_list,diskstars_list,bulgestars_list,m)
File "/ufrc/narayanan/desika.narayanan/pd/source_creation.py", line 244, in add_binned_seds
  #of star particles that go in every [wz,wa,wm]
File "/ufrc/narayanan/desika.narayanan/pd/SED_gen.py", line 225, in allstars_sed_gen
  p = Pool(processes = cfg.par.n_processes)
File "/ufrc/narayanan/desika.narayanan/miniconda2/lib/python2.7/multiprocessing/__init__.py", line 232, in Pool
  return Pool(processes, initializer, initargs, maxtasksperchild)
File "/ufrc/narayanan/desika.narayanan/miniconda2/lib/python2.7/multiprocessing/pool.py", line 159, in __init__
  self._repopulate_pool()
File "/ufrc/narayanan/desika.narayanan/miniconda2/lib/python2.7/multiprocessing/pool.py", line 223, in _repopulate_pool
  w.start()
File "/ufrc/narayanan/desika.narayanan/miniconda2/lib/python2.7/multiprocessing/process.py", line 130, in start
  self._popen = Popen(self)
File "/ufrc/narayanan/desika.narayanan/miniconda2/lib/python2.7/multiprocessing/forking.py", line 121, in __init__
  self.pid = os.fork()
OSError: [Errno 12] Cannot allocate memory

Try reducing the number of n_processes in the parameters_master.py file

5. Assertion Error in SED generation related to zmet being out of range

Error:

assigning stars to SED bins
Running SPS for Binned SEDs
calculating the SEDs for  105  bins
Traceback (most recent call last):
File "pd_front_end.py", line 134, in <module>
m=add_binned_seds(df_nu,stars_list,diskstars_list,bulgestars_list,m)
File "source_creation.py", line 300, in add_binned_seds
binned_stellar_nu,binned_stellar_fnu_has_stellar_mass,disk_fnu,bulge_fnu = sg.allstars_sed_gen(sed_bins_list_has_stellar_mass,diskstars_list,bulgestars_list)
File "SED_gen.py", line 216, in allstars_sed_gen
  add_neb_emission = cfg.par.add_neb_emission, add_agb_dust_model=cfg.par.add_agb_dust_model)
File "build/bdist.linux-x86_64/egg/fsps/fsps.py", line 468, in __init__
  self.params[k] = kwargs.pop(k, v)
File "build/bdist.linux-x86_64/egg/fsps/fsps.py", line 1093, in __setitem__
  self.check_params()
File "build/bdist.linux-x86_64/egg/fsps/fsps.py", line 1067, in check_params
  "zmet={0} out of range [1, {1}]".format(self._params["zmet"], NZ)
  AssertionError: zmet=20 out of range [1, 12]

This is most likely to a mismatch between the isochrones set in parameters_master and the isochrone choice set in sps_vars.f90 in fsps. These should be changed to be the same (and as necessary, recompile fsps and python-fsps).

6. Pool.map errors in powderday

Freezing during Pool.map and ‘metallicity outside of range’ errors:

Entering Pool.map multiprocessing for Stellar SED generation
SSP_GEN ERROR: metallicity outside of range          14
SSP_GEN ERROR: metallicity outside of range          15
...

Some installations have encountered this issue, but its cause has not yet been determined. One potential fix could be using Miniconda instead of Anaconda Python, although this has not been confirmed to be the source of the problem. If something went wrong at any point in the installation process, starting from scratch and doing a fresh installation may also fix the issue.

6. Can’t find “builtins”

During run, we get an error that looks like:

(pd4env) [desika.narayanan@login2 pd_git]$ pd_front_end.py  tests/SKIRT/gizmo_mw_zoom/ parameters_master_401 parameters_model_401
Traceback (most recent call last):
File "/home/desika.narayanan/miniconda3/envs/pd4env/bin/pd_front_end.py", line 4, in <module>
  __import__('pkg_resources').run_script('powderday==0.1.0', 'pd_front_end.py')
File "/home/desika.narayanan/miniconda3/envs/pd4env/lib/python2.7/site-packages/pkg_resources/__init__.py", line 666, in run_script
  self.require(requires)[0].run_script(script_name, ns)
File "/home/desika.narayanan/miniconda3/envs/pd4env/lib/python2.7/site-packages/pkg_resources/__init__.py", line 1469, in run_script
  exec(script_code, namespace, namespace)
File "/home/desika.narayanan/miniconda3/envs/pd4env/lib/python2.7/site-packages/powderday-0.1.0-py2.7.egg/EGG-INFO/scripts/pd_front_end.py", line 7, in <module>

File "build/bdist.linux-x86_64/egg/powderday/__init__.py", line 4, in <module>
File "build/bdist.linux-x86_64/egg/powderday/SED_gen.py", line 19, in <module>
File "build/bdist.linux-x86_64/egg/powderday/nebular_emission/cloudy_model.py", line 2, in <module>
File "build/bdist.linux-x86_64/egg/powderd

  ...

Try the following::

pip install future