Script details
jenkins-server-setup.sh
• This script installs jenkins and other packages like maven, ansible, pip etc.
...
Jenkins setup variable details
ops_ssh_key: This is the private key value using which the VM was created. This is considered as the master key using which we can connect to the VM's
deployer_ssh_key: This is a new key which we generate on local machine or any other machine. Ansible will create anew user (user name will be mentioned in the common.yml file) on all the VM's during bootstrap process. Once this user is created, ansible will use this key for all jenkins jobs. The private key content will be copied into this file during Jenkins setup. The public key will be sprayed to all VM's during bootstrap process.
vault-pass: This is the password to decrypt to the files encrypted using ansibile-vault. The best practice is to encrypt the secrets.yml using ansible-vault and push it to the private github repo. When ansible runs, it will checkout this file and decrypt using the vault-pass file. Even if the secrets.yml is not encrypted, it is a must to enter some value in this value. If the file is empty, Ansible will throw an error even when there is no decryption.
Jenkins environment variable details
...
public_repo_branch: This is a unique variable. All the jobs in Jenkins are by default configured to checkout the Jenkinsfile from this variable as ${public_repo_branch}. When a value like release-1.14 is provided to this variable, the Jenkins jobs checkout the Jenkinsfile from release-1.14 branch from the URL configured in the job. If this value is set to refs/tags/release-1.14, then Jenkins will checkout the Jenkinsfile from the tag named release-1.14. This variable can be changed in Jenkins job configurations and a specific branch or tag name can be specified. This is useful when you want to run some jobs from a different branch or tag instead of the value mentioned in this variable.
deploy-conf (Global pipeline libraries section): This is the name of the library which we have used in Jenkinsfiles. When Jenkinsfile has this library name mentioned, it will checkout couple of common libraries from the URL configured in this section. These libraries are required for the Jenkinsfile to run. To avoid writing same code at multiple places, common code is placed in a separate branch and all Jenkinsfile can use this common code by calling it as a function. If this name is changed, ensure the name in Jenksinfile is also changed to the new library name.
Parameters details
Alll build Build jobs
github_release_tag: Specify a tag name here if you want to build from a tag. Example - release-1.14. This will look for a tag named release-1.14 in the repository URL configured in the Jenkins job and checkout the code from this tag. This should not be confused with public_repo_branch. The public_repo_branch is used only to checkout the Jenkinsfile which has all the build logic. Even if public_repo_branch is configured to some tag name, you will still need to provide a tag name in this parameter box when running the build. If this is empty, it will checkout code from the tag specified in public_repo_branch but it will not tag build artifact with the tag name. Instead it will tag it with commit hash which is undesirable when you want to build from tag.
All Build jobs creates an artifact called metadata.json which will have details such as artifact / docker image name and version, on which Jenkins slave it was built.
ArtifactUpload jobs
absolute_job_path: This is the path from where the metadata.json will be copied. The metadata.json will contain important information like name and version of artifact / docker image and on which Jenkins slave it was built. It is not recommended to change this value. All jobs in Jenkins heavily rely on the metadata.json file to obtain the needed information and the path to this metadata.json file is very critical. For all ArtifactUpload jobs, the metadata.json file will be copied from the Build jobs.