aboutsecurity / OpenCNA

OpenCNA (Collection & Normalization & Analysis)

OpenCNA

OpenCNA (Collection, Normalization and Analysis) tool, to collect data from endpoints (actually, we use rastrea2r to do that), normalize (parse) that data and analyze it.

Requirements

In order to make it easy to deploy, we rely on docker:

• Install and start Docker CE (https://www.docker.com/community-edition)

Deployment

Currently, the docker images are NOT in the docker hub. You have to deploy them first:

sh build.sh

That will create the opencna command line tool and a set of docker images. You can see them, by running:

docker images

Running

Currently, there are two commands for opencna:

• normalize: parses the data in the snapshots and create csv files as output. Example:

  opencna normalize -i data/collector/ -o data/normalizer/

  The snapshots in data/collector/ will be parsed and the result stored in data/normalizer/. The folders in -i and -o will be loaded as volumes in docker. Make sure these folders can be volumes.

• analyze: in this case, the functions work with piping. They read from stdin and write to stdout. Examples:

  • web-history:

    cat data/normalizer/webhistory-WIN81/20171126083616/20171126083616-WIN81-webhist-demouser1.csv | opencna analyze web-history -c URL -o naked_IP

  • random-process-name:

    cat data/normalizer/triage-WIN81/20171126084231-WIN81-process-list.csv | opencna analyze random-process-name -c Name -o is_random_name

  • process-uncommon-nsrl:

    cat data/normalizer/triage-WIN81/20171126084231-WIN81-process-list.csv | opencna analyze process-uncommon-nsrl -c Name -o is_uncommon

  In general, for analyzers, the following arguments must be passed:

  • -c: csv column name(s) to process
  • -o: csv column name(s) generated