This repository contains the structure to build and host map.trenchproject.com through AWS and Docker. Hosted projects can be found at: map.trenchproject.com
Live code can be viewed at rstudio.trenchproject.com with login. Further debugging requires ssh-ing into the EC2 instance using a keypair. Within the EC2 instance:
- code location:
cd /srv/shinyapps - print debugging logs:
docker logs shiny - live containiners:
docker ps
This repository contains a build script to install and run a docker image (also located here) that will create and configure an r shiny and rstudio server to host various r shiny apps on an ec2 instance. To host more/ different apps, simply modify the apps pulled by the script and install any additional packages needed by modifying the dockerfile.
First, check whether the EC2 instance is up and running by logging into console.aws.amazon.com, searching for 'EC2' services, and checking under 'Instances.' If no instance exists, create a new Linux 2 EC2 instance (suggest using a medium instance with >=25gb storage and >=4gb memory). If you create it with the security group "R-Docker-Server", you will automatically provide access to relevant ports. You also will need to have already created a free Docker account. Finally, you will need to add or update the subdomains map and rstudio to the desired domain (under advanced setting in Squarespace domains) as A records using the public ipv4 address of the EC2 instance. Now you have an up-and-running EC2 instance.
SSH into the instance using a keypair on the terminal. To do this, you need our lab's security token. The security token can be found in our internal google drive at 'TrEnCh/Rshiny/isaac-uw.pem'. Once the token is downloaded (I store mine in a '.ssh' directory), navigate to it in terminal and secure it with chmod 400 isaac-uw.pem. This has to be done before it is used to SSH into the EC2 instance.
The code to SSH into the instance can be found by clicking on the instance ID on the website, then clicking 'Connect', and navigating to 'SSH Client'. (The current access command is ssh -i "isaac-uw.pem" ec2-user@ec2-54-149-135-84.us-west-2.compute.amazonaws.com, for example).
If you created a new EC2 instance, follow all the below steps. If an EC2 instance already exists and you're rebuilding it, delete all docker containers with docker rm -f $(docker ps -a -q) and then begin at step 4.
Execute the following commands in the EC2 instance:
- Install Git:
sudo yum install git -y - Clone this repository:
git clone https://github.com/trenchproject/R-Docker-Server.git - Make the scripts executable:
chmod +x ./R-Docker-Server/*.sh - Execute the script
build.shwith proper flags to install and configure Docker (This will terminate the ssh session and may take several minutes):./R-Docker-Server/build.sh - Execute the script
containers.shwith proper flags to build, run, and secure R server containers (This may take several minutes):./R-Docker-Server/containers.sh -u [Desired RStudio Username] -p [Desired RStudio Password]
If you recieve errors regarding access keys or the secret key, find these with:
cat ~/.aws/credentials
After the script has finished executing,
- The RStudio server will be available via a browser at rstudio.trenchproject.com
- The RShiny apps are available via a browser at map.trenchproject.com
- To visit a specific RShiny app, navigate to map.trenchproject.com/<my_app_repository_name>
Note: The inbound security group rules associated with the EC2 instance must be modified to include the following protocols in order to allow access to your newly constructed R ecosystem. If you used the R-Docker-Server security group when creating your instance, don't fret it has been done already!
- Port 80 from anywhere (Webserver- RShiny and RStudio)
- Port 443 from anywhere (Webserver- RShiny and RStudio)