Erlang in Production

By Erlang Central | Published: July 28, 2012

One of the reasons I continue to enjoy writing Erlang systems is how easy they are to maintain in Production. This talk will showcase a number of tools and techniques that will help you investigate and debug problems with production systems and allow you to resolve them with zero-downtime live upgrades.
Talk objectives: You’ll learn how to use stock OTP tools like live code upgrade from the shell without appup or relup, how to use etop to find bottlenecks in your system, how to force crashdumps and inspect them offline, how to inspect and maintain your Production cluster.
Target audience: People who want to know more about the basic tools in OTP that they can use today to troubleshoot and fix Production systems.
Slides are available here.


  • Geoff Cant

    Erlang hacker, Senior Engineer on Heroku's Routing Team

    Geoff Cant is an Erlang Hacker from New Zealand with entirely too little free-time to devote to an ever-expanding collection of github projects. Prior to working on routing and logging systems at Heroku, he has  worked on ejabberd for Process One in Paris and SMS gateways for Catalyst in Wellington. His first Erlang system still sails around the south pacific doing(science) when is_process_alive.

    Favourite quote: "That's not right, that's not even wrong."

    Geoff Cant

Follow Erlang Central:

Have an Erlang Question?

Reach out to the Erlang community