Closed nikola-rados closed 2 years ago
@nikola-rados, In the other projects like https://github.com/bcgov/representation-grant-app, schema-spy is manually built and deployed to dev (via tagging) since the image does not change very often. What happens in the build pipeline is the container gets recycled so it regenerates the schema if there were any changes. Having schema-spy only in dev is sufficient for the use case. If no changes are made the APPS
list in src/bc/baseConfig.groovy
then schema-spy will not make it's way to test
or prod
, since the deployment scripts use the APPS
list to determine which images to tag.
It would be best to follow the project's conventions when adding support for recycling schema-spy which would require a simple oc-recycle.sh
script that uses the oc scale
command be added to the ./scripts
folder and the player.sh
script to be updated to call it. Then you can call the new player
script command form the build pipeline.
@WadeBarnes I've added in a stage to the
Jenkinsfile
but I have some questions.I was using the other
schema-spy
projects as reference but they seemed to have a slightly different approach to the jenkins config. Those projects were using theOpenshiftDSL
plugin, but this project wasn't. Does using it here make sense even though it it not used in any of the other stages? If not, how do we set up?On top of that we know that we don't want to be running
schema-spy
onPROD
. The other projects were using build lists to determine which apps get run. This project didn't have that infrastructure but it also seemed to only mention-dev
in theENV
set up. I'm not sure if that means we are only running this forDEV
and don't need to worry about separating thePROD
apps, or if those references are just a means to get some other information. Do we need to set up something similar to the otherschema-spy
projects in terms of separatingPROD
andDEV
?