Fatal error with read_csv for large file on OS X on both #954
Description
I'm attempting to read a 4.5GB csv file with dask. read_csv reproducibly crashes Python, as per below. This seems different to Fatal error when running read_csv #841?
Python 2.7.11 |Continuum Analytics, Inc.| (default, Dec 6 2015, 18:57:58)
Type "copyright", "credits" or "license" for more information.
IPython 4.0.3 -- An enhanced Interactive Python.
? -> Introduction and overview of IPython's features.
%quickref -> Quick reference.
help -> Python's own help system.
object? -> Details about 'object', use 'object??' for extra details.
In [1]: pdb
Automatic pdb calling has been turned ON
In [2]: import dask.dataframe as dd
In [3]: from dask.diagnostics import Pro
Profiler ProgressBar
In [3]: from dask.diagnostics import ProgressBar
In [4]: df = dd. read_csv('bigcsv_utf_dot_comma.csv')
In [5]: ProgressBar().register()
In [6]: df.compute()
[ ] | 0% Completed | 3.4sFatal Python error: GC object already tracked
/Applications/anaconda/envs/python2/bin/python.app: line 3: 5977 Abort trap: 6 /Applications/anaconda/envs/python2/python.app/Contents/MacOS/python "$@"