Corsaro 3 is a re-implementation of a decent chunk of the original Corsaro which aims to be better suited to processing parallel traffic sources, such as nDAG streams or DPDK pipelines.
- libtrace >= 4.0.12 -- download from https://github.com/LibtraceTeam/libtrace
- libwandio >= 4.2.0 -- download from https://github.com/wanduow/wandio
- libyaml -- https://github.com/yaml/libyaml
- libipmeta -- https://github.com/CAIDA/libipmeta
- libzmq3
- libavro
- libtimeseries -- https://github.com/CAIDA/libtimeseries
- libJudy
- tcmalloc -- https://github.com/gperftools/gperftools
- librdkafka
Debian / Ubuntu users can also find packages for libtrace and libwandio at http://cloudsmith.io/~wand/repos/.
Libyaml, libzmq, libavro, libJudy and tcmalloc should be available through your standard OS package repository -- make sure you install the development versions of those libraries. On Debian / Ubuntu, tcmalloc is included in the libgoogle-perftools-dev package.
On macOS, using HomeBrew you'll want something like brew install avro-c zeromq traildb/judy/judy nwoolls/homebrew-xgminer/uthash.
Ubuntu/Debian packages are available from CAIDA's package repository:
curl https://pkg.caida.org/os/ubuntu/bootstrap.sh | bash
# ^ of course, you should inspect this script first
sudo apt install corsaro
Otherwise, standard installation instructions apply:
./bootstrap.sh (if cloned from GitHub)
./configure
make
sudo make install
There are four tools included with Corsaro 3:
- corsarotagger -- captures packets from a libtrace source and performs some preliminary processing (e.g. geolocation). Emits "tagged" packets onto a multicast group for further downstream analysis.
- corsarotrace -- receives tagged packets from corsarotagger and runs one or more of the built-in analysis plugins against that data stream.
- corsarowdcap -- captures packets from a libtrace source and writes them to disk as a set of trace files.
- corsaroftmerge -- merges interim flowtuple avro files produced by corsarotrace into a single file.
If you have installed Corsaro 3 from source via 'make install', these tools will reside in /usr/local/bin/ by default.
Documentation for each individual tool is included in the docs/ directory in the corsaro3 source tree, or will have been installed on your system alongside this README (if you installed corsaro3 via a package).
Old Corsaro used a pile of CLI arguments for configuring each run. This has now been replaced with YAML configuration files.
The YAML required to use the Corsaro 3 tools is pretty simple to write. There are a number of top-level global config options which are simply set by specifying a key value pair using the following format:
key: value
In the case of corsarotrace, the plugins to apply are expressed as a YAML sequence, with the plugin-specific options appearing as key value pairs following the plugin name itself. For instance, the following config segment will run a corsarotrace instance using the flowtuple and dos plugins. The flowtuple plugin will be configured with the 'sorttuples' option set to 'yes'. The dos plugin will be configured with the 'minattackduration' option set to 60.
plugins:
- flowtuple:
sorttuples: yes
- dos:
minattackduration: 60
Note that the indentation and colon placement is important.
An example configuration file for each corsaro tool is included with the Corsaro 3 source code.
Each corsaro 3 tool will accept a logmode option which is used to determine where any log messages produced by the tool end up. There are four logging modes available:
terminal write log messages to stderr
syslog write log messages to syslog (daemon.log)
file write log messages to a file
disabled do not write log messages