Description

At LastPass, we have millions of daily users, and thousands of requests are hitting our endpoints at any second. At this scale it’s mandatory to have detailed insight into our infrastructure and applications.

To achieve this, we utilize Istio as our service mesh solution and leverage its observability options to make troubleshooting live issues faster and more efficient. In our environments we are federating the metrics of Istio’s bundled Prometheus instances up to production-grade ones to have resiliency and long-term storage for our time series. Beside metrics, Istio provides built-in telemetry solutions like distributed tracing, so we can use Jaeger to follow requests through our services and identify performance bottlenecks and design issues. To get a comprehensive view regarding the health of our mesh we also lean on Kiali’s capabilities.

In this presentation, we will cover the main principles behind LastPass’s observability platform, which helps us keep our infrastructure up and running.

Speakers

Krisztian Fekete
Krisztian Fekete

DevOps Engineer
LogMeIn

Krisztian Fekete is a DevOps Engineer at LogMeIn, where he spends most of his time working on Kubernetes clusters and service meshes. He’s currently building observability platforms around these. He contributes to various open source projects, including Jaeger, Istio, Elasticsearch, and Prometheus-related ones.