Mozilla Services - 30 April 2013
Introducing Heka - Data Acquisition and Collection Made EasyHeka simplifies the collection, analysis, and transportation of data across your cluster/server. Counting downloads, keeping track of downloads from Apache logfiles, reporting security issues is handled in a single unified program thats easy to install locally for development and on servers for deployment.
FeaturesMultiple Data Sources Data can be sent directly in via client libraries or using plug-ins that can read logfiles, handle statsd UDP input, or other custom data sources.
Unified Daemon Heka is capable of routing messages like logstash, aggregating counters and submitting them to other sources like statsd, and transporting log messages from various nodes like syslog.
Plugin Architecture
In case existing plug-ins are insufficient, additional plug-ins may be written in Go or Lua (to run in the embedded Lua sandbox). Plug-ins can be utilized for input, decoding unstructured data, filtering, or as outputs.
Performance Written in Go, heka utilizes light-weight goroutines for fast, efficient message routing, and parallel plug-in operation that can utilize multi-core systems to move hundreds of thousands of messages per second while using a very modest amount of memory.
Usability A single static binary with a library module is all that’s needed to run heka locally, making it easy to develop with and deploy.
Common Uses Application performance metrics
Server load, memory consumption, and other machine metrics
Database, Cache server, and other daemon metrics
Various statsd counters
Log-file transportion and statsd counter generation
Getting Started Install hekad.
Install one of the client libraries or use hekad directly as a message router for various types of messages and/or reading logfiles.
Read the hekad documentation.
Support / Help
Heka developers can be reached via the services-dev mailing list or via IRC on the #heka channel of irc.mozilla.org.
http://heka-docs.readthedocs.org/en/latest/