Setting up RoR/Hydra Environment - Problems and Solutions

  |   Source

Ubuntu 15.10

If you want to setup your RoR environment on Ubuntu 14/15, this post is the best choice for you. I used rvm to install Ruby on my two Ubuntu 15.10 machine, and it worked like a charm.


When I tried to run \curl -L | bash -s stable --ruby in terminal to install both RVM and Ruby, I got following error message:

Error running 'requirements_debian_update_system ruby-2.2.1',
showing last 15 lines of /usr/local/rvm/log/1439385591_ruby-2.2.1/update_system.log
++ /scripts/functions/logging : rvm_pretty_print()  78 > case "${TERM:-dumb}" in
++ /scripts/functions/logging : rvm_pretty_print()  81 > case "$1" in
++ /scripts/functions/logging : rvm_pretty_print()  83 > [[ -t 2 ]]
++ /scripts/functions/logging : rvm_pretty_print()  83 > return 1
++ /scripts/functions/logging : rvm_error()  117 > printf %b 'There has been error while updating '\''apt-get'\'', please give it some time and try again later.
404 errors should be fixed for rvm to proceed. Check your sources configured in:
There has been error while updating 'apt-get', please give it some time and try again later.
404 errors should be fixed for rvm to proceed. Check your sources configured in:
++ /scripts/functions/requirements/ubuntu : requirements_debian_update_system()  53 > return 100
Requirements installation failed with status: 100.

The reason of this error is because RVM will bring error if my Ubuntu machine's apt-get update has error in it. I used to install Ninja-IDE through apt-get and I had PPA error when I do apt-get update. You can view this stackoverflow post.


A user Avinash Raj posted a great solution in Ubuntu community for removing PPA with errors.

The detail of solution is: create a file called ppa-remove (no .sh extension) as following:

sudo rm /tmp/update.txt; tput setaf 6; echo "Initializing.. Please Wait"
sudo apt-get update >> /tmp/update.txt 2>&1; awk '( /W:/ && /launchpad/ && /404/ ) { print substr($5,26) }' /tmp/update.txt > /tmp/awk.txt; awk -F '/' '{ print $1"/"$2 }' /tmp/awk.txt > /tmp/awk1.txt; sort -u /tmp/awk1.txt > /tmp/awk2.txt
tput sgr0
if [ -s /tmp/awk2.txt ]
  tput setaf 1
  printf "PPA's going to be removed\n%s\n" "$(cat /tmp/awk2.txt)"
  tput sgr0
  while read -r line; do echo "sudo add-apt-repository -r ppa:$line"; done < /tmp/awk2.txt > out
  bash out
  tput setaf 1
  echo "No PPA's to be removed"
  tput sgr0

Run sudo chmod +x ppa-remove after you create this file, then move this file to /usr/bin. Run sudo ppa-remove. Your PPA with errors will be removed and you can then re-run \curl -L | bash -s stable or \curl -L | bash -s stable --ruby to install RVM.

Note: Till Aug 2015, If you run \curl -L | bash -s stable --ruby, you will automatically get Ruby 2.2.1 installed. If you run \curl -L | bash -s stable then install Ruby, you can choose Ruby 2.2.2. These two versions do not have conflict.

Mac OS Yoemite

RoR installation process for Mac OS Yosemite is almost the same as Ubuntu. Make sure you have xcode and Homebrew installed first. This post has very detailed description that you can follow.

Project Hydra

I had a post describing my thoughts about Hydra. In this post I will include detail of installation part.

After you finish RoR setting (make sure you have Bundle installed by RubyGem), you want to generate a basic rails app, then add gem 'hydra', '9.1.0.rc1' to your Gemfile. you can re-run Bundle install to install all necessary ruby gems for Hydra.

If you are creating Hydra head for the first time, run rails generate hydra:install to use Hydra generator. If you done this once and you want to setup this project in a new machine, you should skip this step because your old blacklight files will be overwritten.

You can use same Gemfile to set up Hydra environment in different machines, but make sure you run rake jetty:config and rake db:migrate to prevent possible problem.

You can run Rails.env in rails console to check your current environment.

Problem: Data Immigration for Existing Project

How to transfer previous data (Fedora files, Solr documents and indexes) to your new environment?


Well, I'm working on solution now :) To figure out a way I did lots of experiments (and made a big mess) in my Hydra project. I started a new post about it here.

Comments powered by Disqus