Skip to content

Latest commit

 

History

History
216 lines (173 loc) · 12.1 KB

README.md

File metadata and controls

216 lines (173 loc) · 12.1 KB
Raw

ROGUE Cookbook

A stand-alone chef cookbook for the ROGUE-JCTD project. This cookbook can be used to install and configure all of the ROGUE components.

Requirements

Chef community cookbook requirements.

  • apt
  • database
  • git
  • hostsfile
  • java
  • nginx
  • ohai
  • postgresql
  • tomcat

Attributes

rogue::default

  • node['ssl']= Boolean. Determines whether the application will require SSL, defaults to true.
  • node['create_self_signed_cert']= Boolean. If true will generate a self-signed certificate for the node.
  • node['certs'] = A list of certs (as a file paths) to add to Java, Python and system-wide keystores.
  • node['cert']['org'] = The organization element used to generate a self-signed certificate.
  • node['cert']['org_unit'] = The organization unit used to generate a self-signed certificate.
  • node['cert']['country'] = The country used to generate a self-signed certificate.
  • node['cert']['name'] = The common name used to generate a self-signed certificate, defaults to the node's FQDN.
  • node['cert']['certificate'] = The location of the server's SSL certificate (if not generated by this cookbook).
  • node['cert']['key'] = The location of the server's SSL key (if not generated by this cookbook).
  • node['rogue']['debug']= Boolean. Determines whether applications should be configured in debug mode.
  • node['rogue']['aws_rds'] = Boolean. If true, the cookbook will run logic needed to correctly deploy on Amaozn RDS.
  • node['rogue']['version']= The ROGUE version. Defaults to 1.x.
  • node['rogue']['django_maploom']['url'] - The url to use when using pip to download Django-Maploom.
  • node['rogue']['django_maploom']['auto_upgrade'] - Boolean. Determines whether to auto-upgrade the django-maploom version.
  • node['rogue']['user'] = A hash with the username and password for the ROGUE user.
  • node['rogue']['setup_db'] = Boolean. Dictates whether the rogue::database recipe is executed.
  • node['rogue']['geogig']['branch'] = The geogig branch to during the installation.
  • node['rogue']['geogig']['build_from_source'] = Boolean. If true, geogig will build from source code every provision, if false geogig will be downloaded from the node['rogue']['geogig']['url'] location.
  • node['rogue']['geogig']['global_configuration'] = A hash of hashes which represents geogig global configuration.
  • node['rogue']['geogig']['location'] = The geogig installation location.
  • node['rogue']['geogig']['url'] = The git repository url or zip file for geogig.
  • node['rogue']['geoserver']['use_db_client'] = Boolean. If true, geoserver will use the database client in the geonode-geoserver extension to authenticate users.
  • node['rogue']['geoserver']['url'] = The full url to Geoserver in ROGUE.
  • node['rogue']['geoserver']['base_url'] = Geoserver's URL route in ROGUE.
  • node['rogue']['geoserver']['build_from_source'] = Boolean. If true, geoserver will build from the geoserver_ext folder in the rogue_geonode repository.
  • node['rogue']['geoserver']['data_dir'] = The location of the geoserver data directory.
  • node['rogue']['geoserver']['geowebcache']['url'] = The URL used to download GeoWebCache.
  • node['rogue']['geoserver']['jai']['url'] = The URL used to download JAI.
  • node['rogue']['geoserver']['jai_io']['url'] = The URL used to download JAI IO.
  • node['rogue']['geoserver']['war'] = The url of a geoserver .war file. This is only used if node['rogue']['geoserver']['build_from_source'] is false.
  • node['rogue']['geoserver_data']['url'] = The URL used to download a template geoserver data directory.
  • node['rogue']['geoserver_data']['branch'] = The branch or revision to use when downloading a template geoserver data directory.
  • node['rogue']['geoeserver-exts']['branch'] = The branch or revision to use when downloading the geoserver extensions repository, defaults to '2.4.x'.
  • node['rogue']['geoeserver-exts']['location'] = The geoserver-exts installation location.
  • node['rogue']['geoeserver-exts']['url'] = The git repository url for geoserver-exts.
  • node['rogue']['install_docs'] = If true the ROGUE documentation will be built.
  • node['rogue']['interpreter'] = The path to the ROGUE python interpreter.
  • node['rogue']['ip'] = The IP address used to configure ROGUE applications.
  • node['rogue']['iface'] = The network interface used to obtain the ip. Defaults to eth1 when the vagrant attribute is true, eth0 all other times.
  • node['rogue']['logging']['location'] = The location where ROGUE logs should be stored.
  • node['rogue']['networking']['application']['address'] = The application server's ip address (if static).
  • node['rogue']['networking']['application']['fqdn'] = The FQDN used to configure ROGUE applications.
  • node['rogue']['networking']['application']['gateway'] = The application server's gateway (if applicable).
  • node['rogue']['networking']['application']['hostname'] = The application server's hostname.
  • node['rogue']['networking']['application']['netmask'] = The application server's netmask (if applicable).
  • node['rogue']['networking']['database']['address'] = The database server's address (if static).
  • node['rogue']['networking']['database']['gateway'] = The database server's gateway (if applicable).
  • node['rogue']['networking']['database']['hostname'] = The database server's hostname.
  • node['rogue']['networking']['database']['netmask'] = The database server's netmask (if applicable)
  • node['rogue']['nginx']['locations'] = Key, value pairs used to generate nginx location directives.
  • node['rogue']['postgresql']['user'] = The postgres username. Defaults to 'postgres'.
  • node['rogue']['postgresql']['password'] = The postgres password. Defaults to node['postgresql']['password'].
  • node['rogue']['postgresql']['port'] = The postgres port. Defaults to node['postgresql']['password'].
  • node['rogue']['rogue_geonode']['location'] - The path on the node where ROGUE is installed.
  • node['rogue']['rogue_geonode']['url'] - The git repository to use for ROGUE.
  • node['rogue']['rogue_geonode']['branch'] - The branch to use for ROGUE.
  • node['rogue']['rogue_geonode']['settings'] - ROGUE_GEONODE settings (loaded into local_settings.py).
  • node['rogue']['rogue_geonode']['settings']['ALLOWED_HOSTS'] = An array of hosts allowed to make requests to GeoNode.
  • node['rogue']['rogue_geonode']['settings']['PROXY_ALLOWED_HOSTS'] = An array of hosts that the GeoNode can forward requests to.
  • node['rogue']['ssh']['public_key'] = A public key used for the rogue user to support file synchronization.
  • node['rogue']['ssh']['public_key_remote_file'] = The location where to store the public key on the guest machine.
  • node['rogue']['iface'] = The network interface used to obtain the ip. Defaults to eth1 when the vagrant attribute is true, eth0 all other times.
  • node['unison']['user'] = A hash with the username and password for the unison user.
  • node['nginx']['client_max_body_size'] = The max body size for nginx.

Usage

rogue::database

The database recipe will:

  • Create the GeoNode database
  • Create the GeoNode user
  • Create a PostGIS database for imported geospatial data
  • Add a function to authorize layers, using GeoNode permissions when users visit Geoserver

rogue::default

The default recipe will:

  • install ROGUE Geonode from the master branch into /var/lib/geonode/rogue_geonode.
  • install python dependencies
  • serve the the application through nginx.

rogue::fileservice

The fileservice recipe will:

  • Download the fileservice war and deploy it to Tomcat
  • Create a file-service-store in the Geoserver data directory, if it does not already exist

rogue::geogig

The geogig recipe will:

  • install geogig

rogue::geoserver

The geoserver recipe will:

  • Build Geoserver from the geoserver_ext folder in the ROGUE-GEONODE git repository
  • Deploy the Geoserver war to Tomcat
  • Manually deploy the GeoNode-Geoserver extension .jar file to Geoserver's lib folder
  • Download JAI and JAI/IO

rogue::geoserver_data

The geoserver_data recipe will:

  • Clone a default Geoserver data directory from Github if the Geoserver data directory does not already exist
  • Modify the permissions of the Geoserver data directory to work with the unison file synchronization application.

rogue::java

The java recipe will:

  • Install the Orcale Java JDK

rogue::networking

The networking recipe will:

  • Add the host names for the application and database server to the hostsfile

rogue::nginx

The nginx recipe will:

  • Download and install Nginx from source

rogue::permissions

The permissions recipe will:

  • Create a rogue user account
  • Create a roguecat group for files that Rogue and Tomcat both have permissions on

rogue::postgis

The postgis recipe will:

  • Download and compile PostGIS and its dependencies
  • Create a PostGIS database template

rogue::postgresql

The postgresql recipe will:

  • Set the default pg_hba.conf
  • Set the default postgres password
  • Install PostgreSQL Server and call the PostGIS recipe

rogue::tomcat

The tomcat recipe will:

  • Install tomcat7

rogue::unison

The unison recipe will:

  • Install Unison, ACL, and Keychain
  • Update root's fstab entry to add acl

rogue::stig

The stig recipe will:

  • Do a git pull from the ROGUE-JCTD stig repository.

How To

Below are instructions for common configurations that are not default.

Using an existing database server.

A common configuration request is to separate the database from the application server. Follow the steps below to set up ROGUE to use an existing database server. All of the changes will occur in your dna.json file if you are provisioning without Vagrant or in the Vagrantfile if are using Vagrant.

Using Chef only on the application server:

This installation method will refrain from installing Postgres and PostGIS on the application server, but still use the application server to configure the databases.

Note: The databaserecipe assumes that PostGIS is already installed and that a template database called template_postgis is present. You can override the PostGIS template name using the node['postgis']['template_name'] attribute.

  1. Remove rogue::postgresql from your run list to prevent Chef from installing Postgres on the application server.
  2. Specify the network location for your Postgres server in the node['rogue']['networking']['database']['address'] attribute.
  3. Next, populate your postgres connection parameters for the server using the node['rogue']['postgresql']['password'], node['rogue']['postgresql']['user'], node['rogue']['postgresql']['port'] attributes. By default, the database recipe will attempt to connect as postgres in order to create the correct application usernames, tables, and functions. Make sure that your Postgres server will allow this user to login remotely from the application server using the appropriate configuration settings.

Using Chef on the application and database server:

This method separates the application from the database server but installs and configures Postgres and the GeoNode databases using Chef on the database server.

On the Application server
  1. Remove rogue::postgresql from your run list to prevent Chef from installing Postgres on the application server.
  2. Specify the network location for your Postgres server in the node['rogue']['networking']['database']['address'] attribute.
  3. Set the node['rogue']['setup_db'] attribute to false to disable the database configuration step when provisioning.
On the Database server
  1. Add the rogue::networking, rogue::postgresql and rogue::database recipes to your run list.
  2. If desired, add your pg_hba.conf configuration via the node['postgresql']['pg_hba']. Ensure that the application server will be able to connect.
  3. You will need to add your application server's host to node['postgresql']['config']['listen_addresses'].

Contributing

  • Fork the repository on Github
  • Create a named feature branch (like add_component_x)
  • Write your change
  • Write tests for your change (if applicable)
  • Run the tests, ensuring they all pass
  • Submit a Pull Request using Github