search

redhat-cop / openshift-applier

Used to apply OpenShift objects to an OpenShift Cluster
Apache License 2.0
102 stars 62 forks source link

Best way to run applier on multiple repos/inventories from one playbook? #135

Open bparry02 opened 5 years ago

bparry02 commented 5 years ago

I have several git repositories with a single app in each. I want to aggregate the deployment of each of these repos using the applier. What is the best way to accomplish this?

Can the openshift_cluster_content variable be reset between applier runs from a single playbook?

Related to #80

oybed commented 5 years ago

I believe there are a couple of ways you can attempt to do this through the inventory:

1) use a unique host or group to represent your target repo, then define openshift_cluster_content in each respective host_vars or group_vars. You'd need to create a playbook similar to the one that's already in this repo, but remove the [0] from the target hosts to ensure it utilizes all your targets. IMHO, this is pushing the concept of a "host" a bit, but I don't see why it wouldn't work.

2) I've not tried this myself, but you should be able to use the combine filter with openshift_cluster_content definitions from multiple inventories - then just specify multiple -i during the execution. Note that this requires that the content is unique as it doesn't know how to "merge" the content in a smart way (I believe it's typically last-one-wins).