No one is surprised that the implementation of metrics is part of the compulsory minimum for services development. In Jane's team, everyone strives for fault-tolerance because they make mission-critical services for the company. The team, of course, adds metrics: system and business ones to better understand how things are going with the system. Now Kontur's team is confident that they will respond in time to the notification of a critical change in metrics and immediately understand what went wrong... But it wasn't always like that.
Jane will tell how her team was mistaken in collecting, visualizing and interpreting metrics when they lacked data and what problems can be avoided if you understand how tools for collecting and visualizing metrics work. Kontur uses Graphite, Grafana and Moira, but if you have Prometheus or Zabbix this will be interesting for you too.