Hey guys,

I've some problems in finding the best way to setup my pipelines and hope 
you can and will help me :)
To my setup I can say that I have 6 independent services which has to be 
deployed independent from each other. The stages I currently have in all 6 
Pipelines are Test, Build, Release, Publish.
After Release another Pipeline is triggered to run Smoke tests. Our goal is 
to only be able to manually run publish if the smoke tests succeed.

Ideas:
1. start Stage 4 only if the downstream pipeline succeeded (is this 
possible?)
2. start some test pipeline after release and put publish stage into it's 
own pipeline which is triggered after smoke tests
3. add a smoke tests stage and it's material to all pipelines (I don't like 
this)

For the 2nd Idea I got the problem that it's not possible to identify the 
trigger of the smoke tests via the value stream map. It just shows all 
dependency materials but not which triggered it. Therefore it seems like 
the 2nd idea is out. The 1st idea seem to be not possible with gocd and the 
3rd idea is ugly because adding the git material to all pipelines means 
they are triggered and all services will be released just because the smoke 
tests changed.

So does anyone has another ideas to handle this construct test -> build -> 
release (generic) -> smoke test (generic) -> publish (generic)?

Thanks in advance

Best
Thomas

-- 
You received this message because you are subscribed to the Google Groups 
"go-cd" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to