Social

Home Hacking Setup Your own Hacking Lab !! Meta Lab Linux

Setup Your own Hacking Lab !! Meta Lab Linux

Wednesday, 4 July 2018 Edit

Set up instructions

0. Prereqs

  • These instructions assume an Ubuntu 16.04 OS
  • Update serverAddress variable in pages/assets/helpers.R file with YOUR server address
  • Make sure in file scripts/deploy.sh that the line cd $BASENAME/metalab2 has the correct git repository name

1. Create a passwordless metalab user with sudo privilege

adduser metalab
passwd -d metalab
gpasswd -a metalab sudo
Allow metalab user to run sudo commands without password (so that the automatic build script can run sudo) by opening the sudoers file with the visudo command and add the following line to the end of the file:
metalab ALL=(ALL) NOPASSWD: ALL
Now switch users to the metalab user:
su - metalab

2. Install R

See here for instructions (step 6) and also install the purrr, dplyr, and shiny packages (installing these packages can take several minutes):
sudo su - -c "R -e \"install.packages(c('purrr', 'dplyr', 'shiny'), repos='http://cran.rstudio.com/')\""
For the nloptr R package (that is required by one of the Rmd reports), you need to run the following command:
sudo apt-get install libnlopt-dev

3. Install shiny server

sudo apt-get install gdebi-core
wget https://download3.rstudio.org/ubuntu-12.04/x86_64/shiny-server-1.4.2.786-amd64.deb
sudo gdebi shiny-server-1.4.2.786-amd64.deb
Give metalab user permission to write to the shiny apps folder:
sudo chown metalab -R /srv/shiny-server/

4. Install pandoc

sudo wget https://github.com/jgm/pandoc/releases/download/1.19.2.1/pandoc-1.19.2.1-1-amd64.deb
sudo dpkg -i pandoc-1.19.2.1-1-amd64.deb

5. Clone repository into /home/metalab/ directory

sudo apt-get -y install git
cd /home/metalab
git clone https://metalabgithub:<token>@github.com/metalabgithub/metalab2.git
Make sure to use the correct git repository name in the git clone command.
The is the PAT (personal access token) for the metalabgithub GitHub user. You can create a GitHub Personal Access Token at https://github.com/settings/tokens and make sure to select the "repo" scope.

6. Install all required R packages

This is a script that runs every time the site gets built. Let's run it initially manually, it can take about half an hour. Any subsequent run will be fast because only modified package versions will need to be installed.
cd metalab2/scripts
sudo Rscript packageInstaller.R

7. Set git info

git config --global user.email "metalab-contact@stanford.edu"
git config --global user.name "Metalab Cron"

8. Copy /scripts/deploy.sh script into /home/metalab/ and add permissions to it

cd /home/metalab/metalab2/
cp scripts/deploy.sh ../deploy.sh
sudo chmod u+x ../deploy.sh

9. Set up crontab to automatically deploy the site:

crontab -e
and add the following line: 0 0 * * * /home/metalab/deploy.sh > /home/metalab/daily-backup.log 2>&1

Next steps and useful info

VERY IMPORTANT Make sure you only edit using master branch (or by submitting PRs into master). Do not manually edit gh-pages branch because it will get overwritten.
Any change you make in master will be reflected in the site the next time the script runs (within 24 hours).

Scripts

The scripts/deploy.sh script is executed automatically every day and is responsible for:
  1. Making master branch up to date with remote (GitHub version).
  2. Installing and updating R packages (using the scripts/packageInstaller.R script).
  3. Rendering all HTML files (all pages) and copying shinyapps to the server directory on the machine (using the scripts/main_builder.R script).
  4. Preparing gh-pages branch to push, and pushing it to remote.
You generally do not need to touch the scripts folder.
Every time the script runs, it will generate a log file at /home/metalab/daily-backup.log.

Package management.

All packages that you need to have installed for the Rmarkdown or shiny apps need to be listed in metadata/packageInfo.csv.
File specification:
  • Package - package name
  • Version - version of needed package when installed from CRAN (NA when installed from github)
  • Repo - cran or github
  • RepoName - NA when Repo is cran, <github_user>/<github_repo_name> when Repo is github
Whenever this file gets modified, the next build of the site (within 24 hours) will use the new packages.
If you need to update the version of devtools, dplyr, or purrr, you will need to manually install the new version on the machine by logging in and running, for example
sudo su - -c "R -e \"install.packages('dplyr', repos='http://cran.rstudio.com/')\""
This is because these 3 packages are used in the package management script, and installing them while they're in use may result in unpredictable behaviour.

Metadata folder

The metadata folder contains many list-type data that is used throughout the site. To change it, simply edit the corresponding file in the metadata folder and wait for the site to render again. Use these metadata files whenever you add or edit a dataset, a domain, a shiny app, a tutorial tab, a documentation tab, a person in the about page.

Adding new or editing documentation/tutorial/report tabs

Each tab in these 3 pages is an Rmd file under the documentation, tutorials, and reports folders, respectively. Simply add or edit an Rmd there. If you're adding a new tab or want to change the name of the tab, you need to edit the corresponding metadata file in the metadata folder.

Creating reports - no need to source dashboard/global.R

In the previous version of metalab, all Rmd reports had a line: source("../dashboard/global.R", chdir = TRUE). This is not needed, because the build script that renders the Rmd reports already has the code from that setup script, and its environment is accessible to the reports.

Shiny apps

To add a new shiny app or edit a shiny app, look in the shinyapps folder. Don't forget to edit the corresponding metadata file.
All current shiny apps use a common R file shinyapps/common/global.R. You can access it (source it) using source('../common/global.R'). Variables that are currently accessible from this file:
- project_directory - direct path to `metalab2` repository  
- fields  
- fields_derived  
- datasets_file  
- datasets    
- cached_data  
- avg_month   
- all_data  
- studies   
- subjects   
- datasets
If any new packages are required, update the metadata/packageInfo.csv file.

Download metalab linux :- metalab linux

Related Posts

Subscribe Our Newsletter