Vfrdtyky

I have a rather simple branching strategy set up:

• Master


• Development


• Production

After a pull request is approved from Development, I would like to trigger Jenkins to deploy the branch content to a NAS share, overwritting if necessary. The same for Production, except the NAS location is different.

My question is, out of the following options, what would be the best way to accomplish this?

1. Write a bash script to move the files from the Jenkins workspace to the target location.


2. Somehow use git pull directly from the target location.


3. Access share remotely through a service user that has read+write and execute a git pull from the target. Maybe something like \servernamesharenamerepo?


4. None of the above.

Also, I have presented the NAS shares directly to the VM where Jenkins is running. Is this absolutely necessary or is there a better option? I should also mention that these are just scripts, configuration files, and other flat files... nothing needs to be compiled. The total size of the branch could be between 100MB to 200MB (just in case this is too large and would cause perf problems by going the "git pull" route).

I am just getting into the world of continuous delivery so please forgive my ignorance in advance.

UPDATE: I was able to get something working by using the Publish Over SSH plugin. However, now I am trying to figure out the best way to use this plugin. Is it possible to set up a pipeline with this plugin? Right now, I have a generic Freestyle project but need to be able to deploy my Development branch to my Development NAS when a pull request to Development is approved... and need to be able to deploy my Production branch to my Production NAS when a pull request to Production is approved.

Has anyone done this? If so, did you use a Pipeline, Multibranch Pipeline, or Other?

ANOTHER UPDATE: I was able to figure most of everything out. I created a multibranch pipeline with a custom Jenkinsfile similar to the following:

pipeline {

  agent any

  stages {

    stage('Build') {

      steps {

        sh 'echo "Building project"'

      }

    }

    stage('Test') {

      steps {

        sh 'echo "We can eventually build some unit tests here."'

      }

    }

    stage('Deliver for development') {

      when {

        branch 'development' 

      }

      steps {

        sh 'echo "Run Publish over SSH Plugin to push to development NAS."'

        sshPublisher(

          failOnError: true,

          publishers: [

            sshPublisherDesc(

              configName: 'nas_dev',

              transfers: [

                sshTransfer(

                  sourceFiles: '**/*'

                )

              ]

            )

          ]

        )

      }

    }

    stage('Deploy for production') {

      when {

        branch 'production'  

      }

      steps {

        sh 'echo "Run Publish over SSH Plugin to push to production NAS."'

        sshPublisher(

          failOnError: true,

          publishers: [

            sshPublisherDesc(

              configName: 'nas_prd',

              transfers: [

                sshTransfer(

                  sourceFiles: '**/*'

                )

              ]

            )

          ]

        )

      }

    }

  }

}

The problem I now face is around actually mirroring the workspace to the target NAS location. For example, if I delete a file from my git repo, Jenkins fires, creates a workspace, then copies to NAS via Publish over SSH, the file that was deleted is still in the target NAS location. I was hoping this would work similar to what RSYNC would do but I am not finding that to be the case.

Does anyone know how to accomplish this? Also, there is a persistent directory in the target NAS location that I cannot delete.