I'm running MATLAB R2018b on windows 10. I use the typical startup.m file to set some fairly innocuous groot properties and add some folders containing general tools to the MATLAB path.
Recently, I've included a small script that allows me to add additional folders to the path, depending on which project I am working on. It looks something like this:
fprintf('What are you working on?\n')
fprintf('\t1: Project 1.\n')
fprintf('\t2: Project 2.\n')
switch input('')
case 1
addpath(genpath('***Path1***'))
case 2
addpath(genpath('***Path2***'))
otherwise
end
The issue is, when I include this script in the startup.m file, I am unable to start a parpool using the 'local' profile - the cluster validation tool gets hung up on the "Job Test (createJob)" step. But, I have no issues if I run this script outside of the startup.m file (i.e., after initialization of the MATLAB instance). I realize that this solves the problem, but I'm curious as to why I can't run the script from within the startup.m file. Does anybody have any comments?
For what it's worth, when I halt the call to parpool that gets hung, I get:
Operation terminated by user during
parallel.internal.getJavaFutureResult (line 15)
In parallel.internal.pool.InteractiveClient/waitForSessionCreation (line
647)
sessionFuture, 1,
java.util.concurrent.TimeUnit.SECONDS);
In parallel.internal.pool.InteractiveClient/start (line 354)
obj.Session =
obj.waitForSessionCreation(sessionFuture,
connectionCounter, ...
In parallel.Pool>iStartClient (line 796)
spmdInitialized = client.start(poolType , numWorkers, cluster, ...
In parallel.Pool.hBuildPool (line 585)
iStartClient(client, 'pool', cluster, guiMode,
supportRestart, argsList);
In parallel.internal.pool.doParpool (line 18)
pool = parallel.Pool.hBuildPool(constructorArgs{:});
In parpool (line 98)
pool = parallel.internal.pool.doParpool(varargin{:});
