Logging & Docker in Production

From the last few meetups, we have seen lot of interest about learning a bit more about logging and monitoring solutions for Containers. This time, we are going to cover a bit of both. Treasure Data (https://www.treasuredata.com/) has kindly offered to host the event.

We are looking to solidify the agenda over the next few weeks. The format will be similar to our previous meetups - we'll do a survey of logging & monitoring problems and solutions followed by some demo and lightening talks, if possible.

If you are interested in talking about logging and monitoring or have a cool demo to share, do let us know.

Agenda (in works...)

6:00 PM - 6:30 PM Social / Food & Drinks

6:30 PM - 7:20 PM Logging for production systems by "Sada" Sadayuki Furuhashi

7:30 PM - 8:20 PM Open source recipes to build, ship and run production apps with Docker by Jérôme Petazzoni

Details

Logging for Production Systems

Logging is one of the most requested topic that we all want to hear about. Fortunately, we have one of the field experts giving an introduction of challenges and solutions for logging in production systems. Fluentd is the de-facto logging solution for Docker. Sada will also talk about logging in context of containers.

Speaker: Sadayuki Furuhashi

Sada is the original author of Fluentd (http://www.fluentd.org) and also the author of MessagePack (http://msgpack.org/), an open source object serialization library. Sada is a co-founder Treasure Data, Inc. (http://treasure-data.com/), a cloud-based data warehousing and analytics service. He has been working on production distributed systems for a long time. You can find him on twitter as @frsyuki (http://twitter.com/frsyuki)

Open source recipes to build, ship and run production apps with Docker

Getting started with Docker is relatively straightforward:install the Docker Engine, write Dockerfiles for your application components, and put those components together with a Compose file. At this point, your application can be quickly and reliably deployed on any Docker installation, anywhere, in minutes. Great!

How do we take this to production? How do we scale from the single-node setup to a multi-node, highly-available cluster?

Through a whirlwind of live demos, Jérôme will show how to provision a Swarm cluster with Docker Machine, setup overlay networks, host a local registry, deploy ELK to manage logs, implement load balancing, and finally use the resulting platform to build, ship, and run a demo application built around a microservices architecture.

Speaker: Jérôme Petazzoni

Jérôme works at Docker, where he helps others to containerize all the things. He was part of the team that built, scaled, and operated the dotCloud PAAS, before it became Docker. When annoyed he threatens to replace things with a very small shell script. You can find him on twitter as @jpetazzo (https://twitter.com/jpetazzo).