Installing and configuring R in Linux - 3.14

Data360 Analyze R Installation

Product type
Software
Portfolio
Verify
Product family
Data360
Product
Data360 Analyze
Version
3.14
Language
English
Product name
Data360 Analyze
Title
Data360 Analyze R Installation
Copyright
2024
First publish date
2016
ft:lastEdition
2024-06-06
ft:lastPublication
2024-06-06T14:54:31.909942

The following steps assume that the Server and R are on two separate machines. However, it is also possible to install them on the same machine.

CAUTION:
If R is to be deployed on the same machine as it is important to ensure sufficient system resources (especially RAM) are available to handle the peak processing load.
  1. Install R as per the instructions on https://www.r-project.org/

    The current version of the R node was tested against R 3.3.2-3.el7.x86_64.

    If you choose to compile R from source, configure it with the following options to ensure that the supporting binaries for the R node are available:

    • --enable-R-shlib
    • --with-libpng
    • --with-jpeglib
  2. Start R by navigating to the R installation bin directory and executing the R program.

    An R prompt opens, allowing you to enter R commands.

  3. Install the additional packages that the R node needs to execute by typing the following commands in the R prompt window:
    • install.packages("Rserve")
    • install.packages("jpeg")
    • install.packages("png")

    The current version of the R node was tested against the following package versions (later versions of these libraries may not be compatible with the R node):

    • Rserve 1.7-3
    • png 0.1-7
    • jpeg 0.1-8
    Note: As CRAN packages for Linux are typically distributed as source packages, the Linux server must be provisioned with system libraries that permit the compilation of the prerequisite CRAN packages, e.g. gcc with the libraries for Fortran compilation.
  4. Exit R by typing the command q().
  5. Start the X server on your machine, and ensure the identity that runs R has access to X.

    For most systems, this will mean running xinit as root and then ensuring that the DISPLAY environment variable is set to :0 for the identity that starts Rserve. The identity that starts RServe does not need to be in a graphical environment.

  6. Confirm the location of the R temporary directory. The R node will use this directory to store image files as they are in transit from the R server to the server. While the size of each image should be relatively small, they can add up over time, especially on systems with lots of users. The temporary directory can be set up via the standard R mechanisms (see R's tempdir() help for more information).
  7. Configure RServe as per the RServe instructions at: http://rforge.net/Rserve/doc.html#conf. This is a three-step process:
    • Create a custom Rserv.conf configuration file.
    • Consider the security issues discussed in the section that begins "A note about security". The recommended security steps are important to help prevent unauthorized users from executing arbitrary R code on your R Server.
    • Ensure that the following options are in your Rserv.conf file:
      • encoding utf8
      • interactive no
      • remote enable
  8. Navigate to the R bin directory and start the Rserve server by running the following command, where <Rserv.conf> is a reference to your custom <Rserv.conf> file:

    R CMD Rserve --RS-conf <Rserv.conf>

  9. Install the Rserve client on the Server as per the following steps:
    1. Download the Rserve_1.8-3.tar.gz file from the Rserve website: http://rforge.net/Rserve/.
      Note: Later versions of Rserve may not be compatible with the R node.
    2. Unzip and untar the downloaded file, then navigate to: Rserve/inst/java.
    3. Add the REngine.jar and Rserve.jar files to the following location:

      <Data360 Analyze site directory>/lib/java

  10. Download the R Node Pack from the web page.

    Open and import the R node LNA. See the integrated product help for more information on importing library nodes.

  11. By default, the R node is imported into your My Documents folder.

    If you are using the server edition, you can make the R node available for all users on the server by moving it to the Public Documents folder. From the Directory, click My Documents, then select the R node. Click the menu button in the details panel and select Move to Public Documents.

  12. If you configured R to require authentication, ensure that your R node users are given their connection information to enable them to sign in. If the and R Servers are on different machines, you should provide users with a hostname, and a port if you used a custom port for Rserve.

Once you have completed the above steps, see the R node topic in the integrated product help for information on running the R node.