The transition: Manual => Automated => Distributed monitoring
Submitted by saurabh hirani (@saurabh-hirani) on Friday, 15 January 2016
Everyone talks the benefits of having an automated monitoring system in place - one which can discover
infrastructure components as they are added, monitor them while they are alive and stop monitoring when
they are moved out. But no one has chronicled their journey through the process of automating a manually
maintained monitoring system and showcased their battle scars for others to learn from. That’s what
we are going to talk about.
We moved our monitoring systems from a manually maintained icinga 1.x (nagios compatible) setup to an automated, distributed icinga2 infrastructure. This talk is going to cover the following topics:
- The challenges of monitoring a distributed infrastructure
- A manually maintained nagios setup and its limitations
- Prepping for moving to an automated monitoring setup
- Collating data from multiple host sources
- Progressive cutover from manual to automated monitoring
Curiosity and an itch to automate all things manual.
Currently working as a Senior DevOps engineer at BlueJeans. I enjoy automating infrastructure and contributing to the open source community.