# Alfresco benchmark playbook

  
  This ansible playbook allows one to easily deploy an Alfresco benchmark infrastructure on multiple servers.

  ## Requirements

  ### Software requirement

  The playbook has been written and tested with Ansible 2.2. Some additionnal modules have been added to the playbook:
  
   * ansible-xml by cmprescott (requires python-lxml on remote end)
   * selenium

  ### Architectural requirement
  
  This playbook doesn't deploy SSH keys so please make sure you're using it on nodes that already have the needed SSH keys.

  At the moment the target hosts need to have direct acces to internet as we are using public repos.

  Also all names used in the inventory file need to be fully qualified DNS names resolvable by all hosts.

  ## How to use:

  Edit the _inventory_ file and add your (FQDN) hostnames to the appropriate section. If you don't understand the roles of each server, you should probably start reading [Derek Huley's documentation](https://github.com/AlfrescoBenchmark/alfresco-benchmark/tree/master/docs) about the [Alfresco benchmark framework](https://github.com/AlfrescoBenchmark).
  
  To start using the playbook simply start it with:

  ```
  $ ansible-playbook -i inventory alfbm.yml
  ```
  
  If you're loging in as a normal user and need to "sudo", use the command bellow:
  
  ```
  $ ansible-playbook -i inventory alfbm.yml -b -K
  ```

  If some options need to be changed (e.g. java or tomcat version, etc...), edit the group_vars/all file or the host_vars/<HOSTNAME> if the variable is host specific.

  ## TODO

  
   * Make it possible to download software from the Ansible machine instead of target machine (in case target cannot access internet).
   * Allow Oracle JDK usage

   * Allow non-key based play of the book

   * Attach real VNC server to be able to see what selenium actually doing headless

   * move all artifacts URL in a dict, which contains Alfresco version as keys