Observability in Container Environments

Having container native monitoring and alerting for containerized environments (example k8s cluster) is a key element for having a reliable infrastructure.
And as SREs, we need to measure our performance and keep track of the reliability, security ... of these environments.
In this talk Lars Lütcke, Yannick Kubica and Wadhah Ben Aissa from CLOUDETEER’s SRE team will showcase and highlight different tools used for monitoring and alerting in these environments.

Agenda:

• Introduction to SLI and SLO and underlying their importance.
• Highlighting the Kube-Prometheus-Stack and its components (Prometheus, Alertmanager, Node exporters, kube-state-metrics, Loki ...)
• Defining Prometheus and its pillars in terms of putting in place rules, exporters ...
• Defining Alertmanager and showcasing the way alerts get routed to different receivers.
• Drawing the chain of alerts (Prometheus -> alertmanager -> receivers)
• Defining Grafana and exposing its features (importing data sources, dashboards ...)
• Spotlighting Loki’s feature of extracting metrics from logs stored in it.
• Implementing several Prometheus exporters such as Promitor (Azure monitor metrics scraper) and Prom2teams (alerts forwarder).

----------------------------------------------------------------------------------------------------
The event will be held Online