How to run R on Google Compute Engine without Segmentation faults

Question

I'm trying to use Google's computing engine to run R, though keep running into segmentation faults. Here's what I've done so far:

1) I installed Cygwin & Google Cloud SDK and successfully authenticated to the Google Cloud Platform.

2) I've created a virtual machine using the google supplied "debian-7-wheezy-v20140318" image with a "n1-standard-4" machine.

3) I've installed R (with add'l software) on this virtual machine using the commands

sudo apt-get update
sudo apt-get install -y libopenblas-base openmpi-bin libopenmpi-dev r-base openjdk-7-jre openjdk-7-jdk

4) I've verified the software is installed with the command

apt-cache policy r-base

which returns that v2.15.1-4 is installed.

Each time I try to submit a simple R-script to install some common R packages (via the command "sudo R CMD BATCH --no-save installRpkgs.R"), the output log gives me the error

Segmentation fault

When I ssh into the virtual machine try to run R explicitly, I get the same error as well. I have a feeling I'm missing something. Perhaps I need to specify memory or write permissions. I'm not sure, as I am still new at cloud computing. If anyone could help AT ALL, I'd be so grateful.

Thanks!

EDIT: After some troubleshooting, I managed to determine that the culprit is "libopenblas-base". Not sure why it's causing this, but for now I've just continued without it.

Answer 1

You might install the package 'strace', then run 'sudo strace R CMD BATCH --no-save installRpkgs.R' to see what calls are made and their exit status. Unfortunately I'm not familiar with the issue you're having but strace can be immensely helpful in finding out what the issue is when there is little to no output from the program itself.