London Monitoring Winter Meetup

London Monitoring Winter meetup!

Come along for drinks and pizza from 6.30, the talks will begin at 7.00. Afterwards we will go for a drink at a nearby pub.

Thank you to Yelp for hosting this event, and to SignalFX for sponsoring the drinks and pizza!

Location: Yelp, 12-16 Clerkenwell Rd, EC1M 5PQ

Agenda:

Jamie Buchanan - Platform Reliability Engineer @ Trainline - How North Korea Helped Improve Our Mean-Time-To-Recovery

When a service in production breaks the Mean-Time-To-Recovery is directly related to the toil associated with understanding the true state of our platform. In this talk I will take you on the journey we made to reduce this toil. Our mission to seek the truth will take us through Europe, Russia and North Korea. Along the route we will wrestle Pandas, patch things with Greek sticky tape, question our own sanity and breed a bigger faster brood of deer. I will discuss some monitoring architectural theory about what people choose to monitor and why, the increasing importance of observability, the challenges of free choice versus common tooling across teams, the relatively new space of monitoring aggregation tools, and why you should not just accept the turn-key set up and default metrics of some monitoring tools. The main technologies in focus are Logstash, Elasticsearch, Kibana, Big Panda, and Nginx.

Flavien Raynaud & Piotr Chabierski - Software Engineers @ Yelp - Monitoring Large Infrastructure Changes at Yelp

As Distributed System engineers at Yelp, monitoring is one of the foundations we rely on to develop and deploy some of our major infrastructure systems. In this talk, we will guide you through the iterative process of rolling out our new logging pipeline, using tools like SignalFx, Splunk, custom-built collectors and more.

This toolkit proved essential in performing safe and gradual changes, without our developers even noticing. We will give you examples of how these tools are used in practice; in particular to identify and resolve performance bottlenecks, and prevent production incidents.

If you are interested in speaking at an upcoming event, please contact me at londonmonitoring@fastmail.com or via meetup.com