Skip to content

Application Specific Help

Researchers are able to install software via self-service by clicking the pencil icon to the right of the Launch Desktop button -

For best performance, only install the software you need for this phase of the project. You can easily delete the VM and set it up again when you need different software. After checking the boxes for desired software, click the Install button. You will receive an email confirmation when complete.

Nvivo

Nvivo is available for use in the PNR but you must apply a software license at first time use as well as renew the software license annually.

To obtain a Nvivo license (free for Duke users), go to https://software.duke.edu and select and check-out Nvivo. Once complete, the software license will be displayed to you. When you open Nvivo the first time the application will prompt you to update the license key/

SAS

Setting The Work Library Path in SAS Properties

To avoid the issue of SAS generated temporary files using up all of the storage on your C: drive and causing errors such as: 'insufficient resources'. It's important to read/write all files (including temp) from/to your P: drive.

  • Drag SAS from your Start Menu to your Desktop (going forward you will launch SAS from this shortcut)

  • Right-click the newly created SAS shortcut and select Properties from the dropdown

  • Add: -work “P:\read_write\temp” to the end the text that is already populated in the Target text box

  • Temporary files will now be written to the read_write folder in your P: Drive instead of filling up the space in your C: Drive

Stata

Stata can be installed on SAFER Windows VMs, but will require a license key purchased through https://software.duke.edu

Of Note: you will need to unzip any files before importing to Stata

Programming Support (R, Python)

Several programming environments are provided in the software library for PNR VMs.

Anaconda is a cross-platfrom and cross-language package management solution that ensures package compatiability and environment correctness. It is often used for Python, but also supports R, C/C++, Rust, Go, and more. By using Anaconda you are helping to ensure that your code will be portable across more platforms. Through Anaconda, many packages can be installed automatically using 'conda install'

Setting up Jupyter Notebooks using Anaconda:

By default, Jupyter Notebooks won\'t open in the directory containing sensitive data. However, with a few changes in the settings, you can modify this behavior:

Locate Jupyter Notebooks Shortcut:

Find the shortcut for Jupyter Notebooks. Typically, it\'s on your desktop or in the start menu.

Modify Shortcut Properties:

Right-click on the Jupyter Notebooks shortcut and select \"Properties\".

Shortcut Properties

Changing the \"Start In\" Directory:

In the \'Properties\' window, go to the \'Shortcut\' tab.

Find the \'Start in\' field. By default, it has the value %HOMEPATH%.

Change this value to P:/.

Modify the \"Target\" Field:

Still in the \'Shortcut\' tab, locate the \'Target\' field.

Manually modify the end of its value to P:/.

Modify Values

Apply Changes:

Click on \"Apply\" and then \"OK\".

Launch Jupyter Notebooks:

Now, when you launch Jupyter Notebooks using the modified shortcut, it will open in the P:\directory, allowing you to access and work on sensitive data.

R and R Studio

Even though each PNR Windows VM is automatically set to use a proxy at installation, R still needs proxy settings enabled directly within the applications:

the R profile probably needs to be set. Commands to be used:

Set repository

myrepo = getOption("repos") myrepo["CRAN"] = "http://cran.r-project.org" options(repos = myrepo) rm(myrepo)

Set Proxy settings

Sys.setenv("http_proxy"="http://safer-proxy.oit.duke.edu:3128") Sys.setenv("https_proxy"="http://safer-proxy.oit.duke.edu:3128")

For debugging package installation issues in R, turn on with:

options(internet.info = 0)

to turn off again:

options(internet.info = 2)

Using RStudio:

Launch RStudio

In the RStudio console, input the following commands:

Set repository

myrepo = getOption("repos") myrepo["CRAN"] = "http://cran.r-project.org" options(repos = myrepo) rm(myrepo)

Set Proxy settings

Sys.setenv("http_proxy"="http://safer-proxy.oit.duke.edu:3128") Sys.setenv("https_proxy"="http://safer-proxy.oit.duke.edu:3128")

Package Installation:

Now, you can install your desired packages using the install.packages("YourPackageName") command.

Note:

You'll need to input the above settings every time you start a new R session with this method.

Visual Studio Code

The python extension as well as others may be installed via .visx file import

This extension is able to run in vscode after manually selecting python.exe under c:/tools/Anaconda3. When prompted regarding python interpreter, select the following file - python.exe