» rundeck_job

The job resource allows Rundeck jobs to be managed by Terraform. In Rundeck a job is a particular named set of steps that can be executed against one or more of the nodes configured for its associated project.

Each job belongs to a project. A project can be created with the rundeck_project resource.

» Example Usage

resource "rundeck_job" "bounceweb" {
    name = "Bounce Web Servers"
    project_name = "anvils"
    node_filter_query = "tags: web"
    description = "Restart the service daemons on all the web servers"

    command {
        shell_command = "sudo service anvils restart"
    }
}

» Argument Reference

The following arguments are supported:

  • name - (Required) The name of the job, used to describe the job in the Rundeck UI.

  • description - (Required) A longer description of the job, describing the job in the Rundeck UI.

  • project_name - (Required) The name of the project that this job should belong to.

  • group_name - (Optional) The name of a group within the project in which to place the job. Setting this creates collapsable subcategories within the Rundeck UI's project job index.

  • log_level - (Optional) The log level that Rundeck should use for this job. Defaults to "INFO".

  • allow_concurrent_executions - (Optional) Boolean defining whether two or more executions of this job can run concurrently. The default is false, meaning that jobs will only run sequentially.

  • max_thread_count - (Optional) The maximum number of threads to use to execute this job, which controls on how many nodes the commands can be run simulateneously. Defaults to 1, meaning that the nodes will be visited sequentially.

  • continue_on_error - (Optional) Boolean defining whether Rundeck will continue to run subsequent steps if any intermediate step fails. Defaults to false, meaning that execution will stop and the execution will be considered to have failed.

  • rank_attribute - (Optional) The name of the attribute that will be used to decide in which order the nodes will be visited while executing the job across multiple nodes.

  • rank_order - (Optional) Keyword deciding which direction the nodes are sorted in terms of the chosen rank_attribute. May be either "ascending" (the default) or "descending".

  • preserve_options_order: (Optional) Boolean controlling whether the configured options will be presented in their configuration order when shown in the Rundeck UI. The default is false, which means that the options will be displayed in alphabetical order by name.

  • command_ordering_strategy: (Optional) The name of the strategy used to describe how to traverse the matrix of nodes and commands. The default is "node-first", meaning that all commands will be executed on a single node before moving on to the next. May also be set to "step-first", meaning that a single step will be executed across all nodes before moving on to the next step.

  • node_filter_query - (Optional) A query string using Rundeck's node filter language that defines which subset of the project's nodes will be used to execute this job.

  • node_filter_exclude_precedence: (Optional) Boolean controlling a deprecated Rundeck feature that controls whether node exclusions take priority over inclusions.

  • option: (Optional) Nested block defining an option a user may set when executing this job. A job may have any number of options. The structure of this nested block is described below.

  • command: (Required) Nested block defining one step in the job workflow. A job must have one or more commands. The structure of this nested block is described below.

option blocks have the following structure:

  • name: (Required) Unique name that will be shown in the UI when entering values and used as a variable name for template substitutions.

  • default_value: (Optional) A default value for the option.

  • value_choices: (Optional) A list of strings giving a set of predefined values that the user may choose from when entering a value for the option.

  • value_choices_url: (Optional) Can be used instead of value_choices to cause Rundeck to obtain a list of choices dynamically by fetching this URL.

  • require_predefined_choice: (Optional) Boolean controlling whether the user is allowed to enter values not included in the predefined set of choices (false, the default) or whether a predefined choice is required (true).

  • validation_regex: (Optional) A regular expression that a provided value must match in order to be accepted.

  • description: (Optional) A longer description of the option to be shown in the UI.

  • required: (Optional) Boolean defining whether the user must provide a value for the option. Defaults to false.

  • allow_multiple_values: (Optional) Boolean defining whether the user may select multiple values from the set of predefined values. Defaults to false, meaning that the user may choose only one value.

  • multi_value_delimiter: (Optional) Delimiter used to join together multiple values into a single string when allow_multiple_values is set and the user chooses multiple values.

  • obscure_input: (Optional) Boolean controlling whether the value of this option should be obscured during entry and in execution logs. Defaults to false, but should be set to true when the requested value is a password, private key or any other secret value.

  • exposed_to_scripts: (Optional) Boolean controlling whether the value of this option is available to scripts executed by job commands. Defaults to false.

command blocks must have any one of the following combinations of arguments as contents:

  • description: (Optional) gives a description to the command block.

  • shell_command gives a single shell command to execute on the nodes.

  • inline_script gives a whole shell script, inline in the configuration, to execute on the nodes.

  • script_file and script_file_args together describe a script that is already pre-installed on the nodes which is to be executed.

  • A job block, described below, causes another job within the same project to be executed as a command.

  • A step_plugin block, described below, causes a step plugin to be executed as a command.

  • A node_step_plugin block, described below, causes a node step plugin to be executed once for each node.

A command's job block has the following structure:

  • name: (Required) The name of the job to execute. The target job must be in the same project as the current job.

  • group_name: (Optional) The name of the group that the target job belongs to, if any.

  • run_for_each_node: (Optional) Boolean controlling whether the job is run only once (false, the default) or whether it is run once for each node (true).

  • args: (Optional) A string giving the arguments to pass to the target job, using Rundeck's job arguments syntax.

A command's step_plugin or node_step_plugin block both have the following structure:

  • type: (Required) The name of the plugin to execute.

  • config: (Optional) Map of arbitrary configuration parameters for the selected plugin.

» Attributes Reference

The following attribute is exported:

  • id - A unique identifier for the job.