simp-processgraph

This holds the tool we are using to draw the process graphs

This code allows you to plot the communications between your host and others.

• It uses the ss (socket statistics) command with the -npatuw options -n, --numeric Do now try to resolve service names. -a, --all Display all sockets. -p, --processes Show process using socket. -t, --tcp Display only TCP sockets. -u, --udp Display only UDP sockets. -w, --raw Display only RAW sockets.

• It creates an array of hashes of (sitename, hostname, domainname, localIP, localPort, process, user, peerIP, peerPort, socketUsers), and writes the interim data to a file,

• Then it creates a graph, boxing up site, host, IP, ports, and connecting to destinations. Lines are color-alternated to keep them distinct.

In order to create the .png files, you must have graphviz installed

sudo yum install graphviz graphviz-devel graphviz-ruby

...and to ensure you can see the Ruby libraries, type:

export RUBYLIB=/usr/lib64/graphviz/ruby

Below are the functions available under rake:

rake chmod            # Ensure gemspec-safe permissions on all files
rake clean            # Remove any temporary products
rake clobber          # Remove any generated file
rake default          # default - help
rake help             # help
rake pkg:gem          # build rubygem package for simp-processgraph
rake pkg:install_gem  # build and install rubygem package for simp-processgraph
rake spec             # run all RSpec tests

To run the program, build and install the gem by running $ rake pkg:install_gem

and run it $ processgraph -s [sitename]

or: type in the command below to run it right from the ruby: $ ruby simp-processgraph.rb

The parameters are:

Usage: processgraph [options]


    -h, --help                       Help

    -s, --site  NAME                 Name to associate with your site **(REQUIRED)**

    -i, --input filename NAME        Input file or directory name, properly formatted files will have the .ss filetype, generated from an earlier run

    -o, --output file NAME           Output file or directory name (will look for files in the given directory and subdirectories named *.ss)