Create Staging Environment from Productive System

Hi All,

shortly after the Icinga Camp in Berlin this year (thanks for the by the way!) we decided to change our current setup from one VM with all on it (ido, icinga2, web2, director and some module with databases) to a more distributed environment.
Via puppet we are now setting up one Database Node for IDO and all other databases needed, and also via puppet, a Master / Slave HA Cluster. Thats all fine and working.
After the talk from Max I decided to build up a “staging” environment for that which leads me to my question.
When I want to get “test data” from my Productive Icinga Machine (~600 Host ~6000 Services), what would be the best way. I tried to simply dump the databases on Productive Icinga and restore them to my “fresh” cluster. This worked, but had some funny side effects. (deploying new servers via director from staging to productive)
Is there any recommendation from any of you? I’m sure a lot of people here have done the same. I’m really happy for any suggestion.

Thank you all in advance for your help!


my first question would be what you expect of your staging setup.
Building a small env of around 10 servers or more to test new releases, can be done with random values for your checks

Hi Kevin,
thanks for that, my goal would be to have the same hosts/services in staging as in production. To test possible migrations, all the constants we bring with puppet and test all parts of your icinga environment.
So mostly we will use staging to test our puppet classes, that they are not destroying anything when deploying over production setup.