Heka - Collect, aggregate, and visualize your data

An overview of the long term goals of the heka project can be found on ReadTheDocs.

Heka is still very nascent. This repo contains the heka/pipeline project, which is the engine driving the agent and aggregator portions (both of which run as the hekad daemon) of the overall Heka system, as well as some very (very!) basic heka client code and benchmarking tools.

Heka is written in Go. The easiest way to install Heka is through Heka-Build, which will build the correct version of Go, set up a Go environment, and install all required dependencies. Heka-Build also provides a mechanism for easily integrating external plug-in packages into the generated hekad.

It is also possible to install Heka using Go's default go get mechanism. Unfortunately, Heka won't work with Go 1.0.3, the most recent release at the time of this writing. Heka will work with Go 1.1, once it is released. Heka is tested against Go tip, revision 477b2e70b12d (from Dec 7, 2012); it may or may not work with more recent versions of Go tip. Assuming you have a version of Go that will in fact work with Heka, you can install Heka into your current GOPATH using the following command:

go get github.com/mozilla-services/heka/hekad