GoCD will run the Job on all agents that match the environment and resources specified in the job.

If an agent is missing or lost contact, a job will be scheduled. When the agent comes back on line, the job will be re-scheduled.

Jobs are given a unique name based on the name of the job in the configuration file. So for example, if the job is called ‘deploy’ and you have two agents, you would see jobs like ‘deploy-runOnAll-1’ and ‘deploy-runOnAll-2’.

To enable run on all agents for a job, navigate to the Job settings page in the job configuration

To specify that a job should run on all agents, add the attribute

to the job’s definition (see configuration reference for )

Run ‘X’ instances of a Job

If you want to run multiple instances of the same job configuration you do not have to maintain multiple copies of same job config. You can specify how many instances of job you need & Go will take care of spawing the required number of job instances during scheduling.

This feature is particularly useful for test parallelization. It enables Go users to integrate with other test parallelization tools like etc. to achieve distributed test execution with minimal configuration.

To run ‘x’ instances of a job, navigate to the Job settings page in the job configuration

To specify that ‘x’ instances of a job should run, add the attribute

to the job’s definition (see configuration reference for <job>)

    ...

Also See…

• Re-running job(s)