Add an new option in the infrabox job options list to rerun this job if it failed.
Here is an example infrabox.json file,
{
"version": 1,
"jobs": [{
"type": "docker",
"name": "retest",
"resources": { "limits": { "cpu": 1, "memory": 1024 } },
"build_only": false,
"docker_file": "Dockerfile",
"retry": "1"
}]
}
when "retry" = 0, infrabox works as before.
when "retry" = N, infrabox will rerun the job "retest" N times if the prior execution of this job failed.