AWS_RStudio.Rmd

---
title: How to set up geospatial software in AWS EC2 the right way
author: Manish Verma
output: html_notebook
---

This document describes the steps I followed (with help from my son, who
is a Linux guru) to set up R, RStudioServer, Python, and Jupyter
Notebook for geospatial analysis.

We describe a method below that allows you to set up your own password
and username for logging in to the server.

We first created an Amazon Linux 2 instance. The package maanger,
`amazon-linux-extras`, only supported R 3.4 and we discovered that some
of the most important packages that I wanted such as 'sf' and 'lwgeom'
are not available for 3.4 anymore. So, we terminated the instance and
created a new one with the latest version of Ubuntu ([Ubuntu
20.04](https://aws.amazon.com/marketplace/pp/B087QQNGF1) at the time of
this writing).

You can read the AWS documentation about how to set up an EC2 instance.
When creating the instance, make sure that you configure the Security
Group settings to allow SSH and also open the port 8787 for TCP
connections so that you can connect to RStudio Server via web. RStudio
Server Open Source does not support encrypted connectsions (HTTPS) so be
careful about which IPs you want to allow. Ideally, you want to only
allow the IP of the machine you are using. If it changes you can always
edit the Security Group. Once you are logged into your VM, follow the
steps below to set up the software. You can also open port 8888 if you
want to use Jupyter Notebook, which *does* support HTTPS.

1.  Update your packages.
    ``` {.bash}
    sudo apt update
    sudo apt upgrade
    ```

    You may have to install some C libraries that R packages depend on.
    The following are two libraries that are dependencies of rgdal and
    units:
    ``` {.bash}
    sudo apt install libudunits2-dev libgdal-dev
    ```

2.  Install and configure RStudio Server.
    1.  Install R and RStudio Server according to the insturctions from
        the RStudio website.
        ``` {.bash}
        sudo apt install r-base # should get you R 3.6.3)
        sudo apt install gdebi-core
        wget --no-verbose https://download2.rstudio.org/server/bionic/amd64/rstudio-server-1.2.5042-amd64.deb
        sudo gdebi rstudio-server-1.2.5042-amd64.deb
        ```

        After the last command, RStudio Sever (the binary is called
        `rserver`) should be started automatically. If you want to make
        sure you can run `ps -aux` to see a list of all processes that
        are currently running.

    2.  If you were to navigate to your EC2 instances public ip right
        now, there would be one main issue: you would not be able to log
        in because your default account (`ubuntu`) does not have a
        password.  Set a password for your default account by running
        ``` {.bash}
        sudo passwd ubuntu
        ```

    3.  Now reload the server so it updates its configuration
        ``` {.bash}
        sudo systemctl restart rstudio-server.service
        ```

    4.  Navigate to `http://{EC2 instance's public ip}:8787` in your web
        browser and enter the username and password for your default
        account.

3.  Setting up Jupyter Notebook is somewhat easier.

    1.  Install Anaconda Python (or whatever distribution you prefer)
        ``` {.bash}
        wget -nv https://repo.anaconda.com/archive/Anaconda3-2020.02-Linux-x86_64.sh
        bash Anaconda3-2020.02-Linux-x86_64.sh
        ```

    2.  Create a self-signed SSL certificate by running
        ``` {.bash}
        mkdir ~/ssl
        cd ~/ssl
        openssl req -x509 -nodes -days 365 -newkey rsa:4096 -keyout mykey.key -out mycert.pem
        ```

        This allows you to log in to the RStudio Server from your web
        browser using HTTPS instead of HTTP.

    3.  Create a Jupyter Notebook password by running
        ``` {.bash}
        jupyter notebook password
        ```

    4.  Start the server and connect to it. First, run the following on
        your EC2 instance:
        ``` {.bash}
        jupyter notebook --certfile=~/ssl/mycert.pem --keyfile ~/ssl/mykey.key
        ```

        Then, run the following on your local machine to forward your
        local port 8888 to port 8888 on your EC2 instance:
        ``` {.bash}
        ssh -i {EC2 private key} -NfL 8888:localhost:8888 ubuntu@{EC2 instance's public ip}
        ```

        Finally, visit <https://localhost:8888> and log in.