Servcie cluster and severity

hello guys,

Tell me please what is the best practice to provide accurate alerts and notifications for microservices. When they are working in clusters.

here is an example:

  • we have a microservice called “validation123”.
  • he resides on 4 nodes, Linux systems.
  • so, in the end, we have 4 running daemons.
  • if one “validation123” service is down I want to have a warning status with PRIO 3 notification.
  • if two “validation123” services are down I want to have a warning status with PRIO 2 notification.
  • if one “validation123” service is down I want to have a warning status with PRIO 3 notification.
  • if three or four “validation123” services are down I want to have a critical status with PRIO 1 notification.

Maybe, Icinga Business Process Modelling + icingacli-businessprocess? Would it generate an alert notification?

Marcin

Seems that I found a ready-to-use solution.
https://community.icinga.com/t/checking-service-clusters-in-icinga-2/752 (github.com)