Creating 2 pipelines in a single jenkinsfile

[Faad Sayaou]

Hi everyone,

I have this scenario i would like to implement but i don not know if that is possible within a single jenkinsfile.

My idea was to have 2 pipelines

1st pipeline

 -git checkout

 -build

 -Test

2nd pipeline

if first build from the first pipeline was successful, then

-re-do the first 3 stages from 1st pipeline

-zip the build results

-and push to nexus

..is this however feasible within a single jenkinsfile?

thanks

[Faad Sayaou]

I was thinking maybe doing something like this would help but unfortunately not

node {

stages....

}

node2 {

stages....

}

but it is viewed by jenkins as a single pipeline. I am using the scripted pipeline syntax

[Denis]

You can create two pipeline jobs and have the first one trigger the second one as a downstream job.
i.e in a finally block you can check the result of the build and trigger the pipeline that publish to nexus

--
Sent from my Android device with K-9 Mail. Please excuse my brevity.

[Arnaud bourree]

Hi,

Denis answer is good but why are you doing that?

You said if pipeline #1 succeed, you run pipeline #2 which redo same steps including git checkout then push result to nexus. 

But you don't have guarantee that git checkout will fetch same source code. 

Why not add input validation on nexus stage in single pipeline?

Arnaud 

--
You received this message because you are subscribed to the Google Groups "Jenkins Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-users/97F68F8F-DB51-4C09-B894-D8A57A30DFB0%40gmail.com.
For more options, visit https://groups.google.com/d/optout.

[Adrien Lecharpentier]

I agree with Arnaud. 

However, having an input means that if you cancel the input, you'll have aborted the build. So, even if the first 3 steps are successful, you'll have an aborted build.

You could try/catch the input, but that will complexify the pipeline imo.

Could you explain why you need to conditionally upload the results and not doing it each time it's successful? If you have some concerns about the "space usage", you could make sure to replace the same package each time, or configure nexus to have some cleanup tasks. 

To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-users/CAESUf_AYVtmQN%2BhinWSB-DgLtZ3QKTHYn0woYeEHQLfEwiBw4Q%40mail.gmail.com.

For more options, visit https://groups.google.com/d/optout.

--

Adrien Lecharpentier