OpenNMS Configuration Tools v0.0.2

WARNING: those configuration tools are on an alpha stage, don't use them on 
production environment. No really, don't!

They have been written to allow an easier and faster integration of OpenNMS on a
small organisation with only some simple tasks to do. OpenNMS is much more than
that and those tools covers only a small part of what it can do. One can think
of them as something who simplify the first steps of OpenNMS integration.

Tested on Debian 5.0 Lenny and Centos 5.4.


[install.py]
This tool automates the installation of OpenNMS for Debian and RedHat based
distributions. Optionaly one can chose between the stable and developement 
version, see "install.py --help" for usage details.
It takes the following steps:
    1. Determination of the current distribution
    2. Warn if a previous version have been detected
    3. Add OpenNMS packages repository
        3.1. Sign the repository on Debian
    4. Installation of OpenNMS packages and his dependencies
        4.1. Let PostgreSQL populating his configuration directory on RedHat
        4.2 Modification of PostgreSQL's configuration for OpenNMS
    5. Creation of OpenNMS database
    6. Installation of IPLike
    7. Configuration of Java Runtime Environment
    8. Installation of OpenNMS


[config.py]
This tool is one of the main reason for creating this suite. The configuration
of OpenNMS must be the result of a discution concerning the ojectives of the 
monitoring plateform, especialy for the following points:
    * Who are the users of such plateform and what are their roles and 
      credentials?
    * What are the IP addresses of the network equipments we want to monitor?
    * What are the credentials needed by some protocols to access those 
      equipments?
    * Is there are specific tasks we want to do with the help of some plugins?

All those rules are summarized on a manifest file named [config_rules.py] which
is used by this tool to automate the configuration of OpenNMS.
"config.py --save" will apply a configuration and "config.py --remove" will
remove the plugin listed in [config_rules.py].

WARNING: allways run "backup.py" before any changes and be aware that 
"config.py --remove" will only work if the plugin files have not been changed.

A standard usage should takes the following steps:
    1. Execution of "backup.py"
    2. Execution of "config.py --remove"
    3. Modification of [config_rules.py] and/or [plugin/*.py]
    4. Execution of "config.py --save"

The purpose of that tool is to facilitate the upgrade process.


[plugin/plugin.py]
This file is called by [config.py] for activating or desactivating each plugin
designed by [config_rules.py]. It makes [config.py] very usefull when you want
to easily keep track of all the modifications you've done tho OpenNMS 
configuration files. 


[config_rules.py]
This file is a manifest of OpenNMS configuration, see [config.py].


[plugin/apache.py]
An integration example for Apache HTTPd Server, it contains all the 
modifications that must be done to OpenNMS configuration file.


[plugin/nginx.py]
Same thing for Nginx HTTPd Server.


[backup.py]
This tool archives OpenNMS configuration files and all his other files too 
but the logs files into an unique sub-directory of a given directory. It also 
dump the database in it.


[restore.py]
This tool is the counterpart of [restore.py]. It will overwrite the current
configuration state with the previous state stored in the given unique 
sub-directory where it has been saved by [backup.py].


[test.py]
This tool contains a list of test cases to see if OpenNMS is on a consistent
state. Those are the following:
    1. OpenNMS packages and his dependencies are installed on the system
    2. IPLike and PostgreSQL Server are version compatible
    3. OpenNMS is running
    4. PostgreSQL Server is running
    5. The plugins' tools are installed

This tool also contains some regression tests to check if nothing have changed
between the current state and the given previous one. A state can be saved at 
any time by running "test.py --save". The regression tests are 
the following:
    1. Nodes seen within the Web GUI
    2. Interfaces seen within the Web GUI
    3. Users seen within the Web GUI
    4. Groups seen within the Web GUI
    5. Nodes seen within the database
    6. Interfaces seen within the database
    7. Users declared in the XML files
    8. Groups declared in the XML files


[lib/distrib.py]
This library contains some functions with peculiar code for each distribution.


[lib/system.py]
This library contains some (ugly) functions to wrap some specific interactions
with the system done by the tools.