R help - Dec 2015 - Regular R freezes

Dear List,

some days ago my R started to regularily freezes, especially during 
these operations:

mso(,permutations=999), package vegan
anova.cca(), package vegan
forward.sel(), package packfor

I am running R on a 64-bit Windows 7 installation, 4 cores, 8 GB RAM.
I just today updated R to 3.2.3, all packages, and R-Studio to their 
most recent versions.
Still the error persists.
Two things are strange:

All these functions (and more) have been working without any problem for 
2 years.
The freezes occur at different points in the algorithms. For 
forward.sel(), a stepwise variable selection process
in species distribution modelling, the system fails at different steps 
(say after variable 8 or 18), but inevitably fails.
R-Studio is still responsive meanwhile, but the "Stop"-Buttom does not
work, and the programm needs to be restarted.

Is there any way of troubleshooting or finding out who the culprit is?

Thanks.

-- 
Tim Richter-Heitmann (M.Sc.)
PhD Candidate



International Max-Planck Research School for Marine Microbiology
University of Bremen
Microbial Ecophysiology Group (AG Friedrich)
FB02 - Biologie/Chemie
Leobener Stra?e (NW2 A2130)
D-28359 Bremen
Tel.: 0049(0)421 218-63062
Fax: 0049(0)421 218-63069

On Tue, 22 Dec 2015 16:13:07 +0100
Tim Richter-Heitmann <trichter at uni-bremen.de> wrote:
> Dear List,
> 
> some days ago my R started to regularily freezes, especially during 
> these operations:
> 
> mso(,permutations=999), package vegan
> anova.cca(), package vegan
> forward.sel(), package packfor
> 
> I am running R on a 64-bit Windows 7 installation, 4 cores, 8 GB RAM.
> I just today updated R to 3.2.3, all packages, and R-Studio to their 
> most recent versions.
> Still the error persists.
> Two things are strange:
> 
> All these functions (and more) have been working without any problem
> for 2 years.
> The freezes occur at different points in the algorithms. For 
> forward.sel(), a stepwise variable selection process
> in species distribution modelling, the system fails at different
> steps (say after variable 8 or 18), but inevitably fails.
Look at how much memory is available.  Possibly increase the work space
available to R.  What other software was recently changed immediately
prior to the appearance of the problem.  Lastly, check that hardware it
self.  Most hard drives have diagnostic routines that can be accessed
to check drive integrity.  

The random nature to the fail suggests a memory availability problem -
some newly resident routine that uses more memory than it should.  Run
the Task Manager in monitor memory availability before, and at
failure.  But, I once chased a similar problem all the way to a failing
CPU before fixing it. I essentially rebuilt the computer.  

> R-Studio is still responsive meanwhile, but the "Stop"-Buttom
does
> not work, and the programm needs to be restarted.
> 
> Is there any way of troubleshooting or finding out who the culprit is?
> 
> Thanks.
> 
JWDougherty