# Rundeck

> Run your Rundeck runbooks: run jobs with options and wait for them, run commands on nodes, find, follow and abort executions and read their logs, turn jobs and schedules on and off, import jobs, keep Key Storage up to date, and start a workflow when a job fails, succeeds or starts.

App · 20 actions. Web page: https://docs.lodol.com/docs/integrations/rundeck

## Actions

### Abort execution

Stop a running execution, as its Kill Job button does, and wait a few seconds for it to stop. One that has already stopped is left as it is.

**Inputs**

- Execution (`execution_id`, text, required): The execution's number, as Rundeck shows it (#123), or a link to it. 'Run job' and 'Run command' give it as id.

**Outputs**

- `abort_status` (text): What Rundeck said to the request: aborted or pending; empty if it had already stopped
- `argstring` (text): The options it ran with, as Rundeck writes them
- `changed` (true/false): Whether it was running and asked to stop
- `custom_status` (text): The status a job set itself, when status is other
- `description` (text): What it ran: the job's steps, or the command
- `duration_seconds` (number): How long it ran, or has run so far, in seconds
- `ended` (text): When it stopped (ISO 8601); empty while it runs
- `execution_type` (text): user (by a person or the API), scheduled or user-scheduled
- `failed_nodes` (list): The nodes it failed on
- `finished` (true/false): Whether it has stopped
- `id` (number): The execution's number, as Rundeck shows it (#123)
- `job_full_name` (text): The job's group and name; empty for a command
- `job_group` (text): The job's group; empty if none
- `job_id` (text): The job's ID; empty for a command
- `job_name` (text): The job's name; empty for a command
- `link` (text): A link to the execution in Rundeck
- `options` (object): The options it ran with, by name
- `project` (text): The project it ran in
- `started` (text): When it started (ISO 8601); empty until it does
- `status` (text): running, succeeded, failed, aborted, timedout, failed-with-retry, scheduled (set to run later) or other (a status the job set itself)
- `succeeded` (true/false): Whether it succeeded
- `successful_nodes` (list): The nodes it succeeded on
- `user` (text): Who started it

**Example output**

```json
{
  "abort_status": "pending",
  "argstring": "-environment staging",
  "changed": true,
  "custom_status": "",
  "description": "systemctl restart web",
  "duration_seconds": 12.4,
  "ended": "2026-03-02T10:15:12Z",
  "execution_type": "user",
  "failed_nodes": [],
  "finished": true,
  "id": 123,
  "job_full_name": "deploy/web/restart",
  "job_group": "deploy/web",
  "job_id": "9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
  "job_name": "restart",
  "link": "https://rundeck.example.com:4440/project/ops/execution/show/123",
  "options": {
    "environment": "staging"
  },
  "project": "ops",
  "started": "2026-03-02T10:15:00Z",
  "status": "aborted",
  "succeeded": false,
  "successful_nodes": [
    "web1"
  ],
  "user": "lodol"
}
```

### Delete execution

Delete a finished execution and its log from Rundeck's history. One that isn't there is left as it is; one still running must be aborted first.

**Inputs**

- Execution (`execution_id`, text, required): The execution's number, as Rundeck shows it (#123), or a link to it. 'Run job' and 'Run command' give it as id.

**Outputs**

- `deleted` (true/false): Whether it was deleted; false if Rundeck had no such execution
- `id` (number): The execution's number
- `status` (text): Its status before it was deleted

**Example output**

```json
{
  "deleted": true,
  "id": 123,
  "status": "succeeded"
}
```

### Delete job

Delete a job, as its menu's Delete this Job does. Its past executions stay. A job that isn't there is left as it is.

**Inputs**

- Job (`job_id`, text, required): The job: pick one, or paste its ID or a link to its page in Rundeck.

**Outputs**

- `deleted` (true/false): Whether it was deleted; false if Rundeck had no such job
- `id` (text): The job's ID
- `name` (text): Its group and name; empty if Rundeck had no such job
- `project` (text): The project it was in

**Example output**

```json
{
  "deleted": true,
  "id": "9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
  "name": "deploy/web/restart",
  "project": "ops"
}
```

### Delete key

Delete a password or key from Key Storage. Jobs and nodes that use it stop working until it's stored again. A path with nothing there is left as it is.

**Inputs**

- Path (`path`, text, required): The key's path, such as keys/project/ops/db.password.

**Outputs**

- `deleted` (true/false): Whether it was deleted; false if there was nothing there
- `path` (text): The key's path

**Example output**

```json
{
  "deleted": true,
  "path": "keys/project/ops/db.password"
}
```

### Enable or disable job

Turn a job's running or its schedule off or on, as the job's menu in Rundeck does: with running off, nobody can run it; with its schedule off, it runs only when started by hand.

**Inputs**

- Job (`job_id`, text, required): The job: pick one, or paste its ID or a link to its page in Rundeck.
- Turn (`part`, text): What to turn off or on. One of: `execution` (Running it (by hand, by schedule or by the API)), `schedule` (Its schedule only). Default: `execution`.
- On (`enabled`, true/false, required): On to enable it, off to disable it.

**Outputs**

- `changed` (true/false): Whether it changed; false if it already was
- `description` (text): Its description; empty if none
- `execution_enabled` (true/false): Whether it may run at all (by hand, by schedule or by the API)
- `full_name` (text): Its group and name, such as deploy/web/restart
- `group` (text): The group (folder) it's in, such as deploy/web; empty if none
- `id` (text): The job's ID (a UUID), as its page's address shows it
- `link` (text): A link to the job in Rundeck
- `name` (text): The job's name
- `part` (text): What was turned on or off: execution or schedule
- `project` (text): The project it's in
- `schedule_enabled` (true/false): Whether its schedule is on
- `scheduled` (true/false): Whether it has a schedule

**Example output**

```json
{
  "changed": true,
  "description": "Restart the web service",
  "execution_enabled": true,
  "full_name": "deploy/web/restart",
  "group": "deploy/web",
  "id": "9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
  "link": "https://rundeck.example.com:4440/project/ops/job/show/9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
  "name": "restart",
  "part": "schedule",
  "project": "ops",
  "schedule_enabled": false,
  "scheduled": true
}
```

### Find executions

Find a project's executions, the latest first, by status, job, who started them and when they ran, as the Activity page does.

**Inputs**

- Project (`project`, text, required): The project, by its name as Rundeck lists it, such as ops.
- Status (`status`, text): Only executions with this status. One of: `any` (Any status), `running`, `succeeded`, `failed`, `aborted`, `timedout` (Timed out). Default: `any`.
- Job (`job_id`, text): Only executions of this job.
- Started by (`started_by`, text): Only executions this Rundeck user started.
- Since (`since`, date and time): Only executions that ran at or after this time.
- Until (`until`, date and time): Only executions that ran at or before this time.
- How many (`limit`, number): The most to list, 1 to 200. Defaults to 20. From 1 to 200.

**Outputs**

- `count` (number): How many are listed
- `executions` (list): The executions, the latest first
- `project` (text): The project
- `total` (number): How many match, listed or not

**Example output**

```json
{
  "count": 1,
  "executions": [
    {
      "argstring": "-environment staging",
      "custom_status": "",
      "description": "systemctl restart web",
      "duration_seconds": 12.4,
      "ended": "2026-03-02T10:15:12Z",
      "execution_type": "user",
      "failed_nodes": [],
      "finished": true,
      "id": 123,
      "job_full_name": "deploy/web/restart",
      "job_group": "deploy/web",
      "job_id": "9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
      "job_name": "restart",
      "link": "https://rundeck.example.com:4440/project/ops/execution/show/123",
      "options": {
        "environment": "staging"
      },
      "project": "ops",
      "started": "2026-03-02T10:15:00Z",
      "status": "succeeded",
      "succeeded": true,
      "successful_nodes": [
        "web1"
      ],
      "user": "lodol"
    }
  ],
  "project": "ops",
  "total": 1
}
```

### Get execution

Get an execution's status, options, times and nodes, with its state on each node.

**Inputs**

- Execution (`execution_id`, text, required): The execution's number, as Rundeck shows it (#123), or a link to it. 'Run job' and 'Run command' give it as id.

**Outputs**

- `argstring` (text): The options it ran with, as Rundeck writes them
- `custom_status` (text): The status a job set itself, when status is other
- `description` (text): What it ran: the job's steps, or the command
- `duration_seconds` (number): How long it ran, or has run so far, in seconds
- `ended` (text): When it stopped (ISO 8601); empty while it runs
- `execution_type` (text): user (by a person or the API), scheduled or user-scheduled
- `failed_nodes` (list): The nodes it failed on
- `finished` (true/false): Whether it has stopped
- `id` (number): The execution's number, as Rundeck shows it (#123)
- `job_full_name` (text): The job's group and name; empty for a command
- `job_group` (text): The job's group; empty if none
- `job_id` (text): The job's ID; empty for a command
- `job_name` (text): The job's name; empty for a command
- `link` (text): A link to the execution in Rundeck
- `node_states` (list): How it went on each node
- `options` (object): The options it ran with, by name
- `project` (text): The project it ran in
- `started` (text): When it started (ISO 8601); empty until it does
- `state` (text): Its state in detail: WAITING, RUNNING, RUNNING_HANDLER, SUCCEEDED, FAILED, ABORTED, NODE_PARTIAL_SUCCEEDED or NODE_MIXED
- `status` (text): running, succeeded, failed, aborted, timedout, failed-with-retry, scheduled (set to run later) or other (a status the job set itself)
- `succeeded` (true/false): Whether it succeeded
- `successful_nodes` (list): The nodes it succeeded on
- `target_nodes` (list): The nodes it runs on
- `user` (text): Who started it

**Example output**

```json
{
  "argstring": "-environment staging",
  "custom_status": "",
  "description": "systemctl restart web",
  "duration_seconds": 12.4,
  "ended": "2026-03-02T10:15:12Z",
  "execution_type": "user",
  "failed_nodes": [],
  "finished": true,
  "id": 123,
  "job_full_name": "deploy/web/restart",
  "job_group": "deploy/web",
  "job_id": "9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
  "job_name": "restart",
  "link": "https://rundeck.example.com:4440/project/ops/execution/show/123",
  "node_states": [
    {
      "node": "web1",
      "state": "SUCCEEDED"
    }
  ],
  "options": {
    "environment": "staging"
  },
  "project": "ops",
  "started": "2026-03-02T10:15:00Z",
  "state": "SUCCEEDED",
  "status": "succeeded",
  "succeeded": true,
  "successful_nodes": [
    "web1"
  ],
  "target_nodes": [
    "web1"
  ],
  "user": "lodol"
}
```

### Get execution output

Get the end of an execution's log, oldest line first, with the lines logged as errors, from every node or one.

**Inputs**

- Execution (`execution_id`, text, required): The execution's number, as Rundeck shows it (#123), or a link to it. 'Run job' and 'Run command' give it as id.
- How many lines (`lines`, number): The most lines to get, the last ones, 1 to 2000. Defaults to 100. From 1 to 2000.
- Node (`node`, text): Only the lines from this node, by its name. Leave blank for every node.

**Outputs**

- `count` (number): How many lines there are
- `errors` (list): The lines logged as errors
- `execution_id` (number): The execution's number
- `finished` (true/false): Whether it has stopped, so the log is whole
- `lines` (list): The last lines, oldest first
- `log` (text): The lines, as one text
- `status` (text): Its status, such as running or succeeded

**Example output**

```json
{
  "count": 2,
  "errors": [],
  "execution_id": 123,
  "finished": true,
  "lines": [
    {
      "level": "NORMAL",
      "message": "Restarting web",
      "node": "web1",
      "step": "1",
      "time": "2026-03-02T10:15:01Z"
    },
    {
      "level": "NORMAL",
      "message": "web is running",
      "node": "web1",
      "step": "1",
      "time": "2026-03-02T10:15:11Z"
    }
  ],
  "log": "Restarting web\nweb is running",
  "status": "succeeded"
}
```

### Get job

Get a job's definition: its options (what a run can be given, which are required and their defaults), schedule, nodes and steps, and, if asked, its YAML to keep or import elsewhere.

**Inputs**

- Job (`job_id`, text, required): The job: pick one, or paste its ID or a link to its page in Rundeck.
- Include YAML (`include_yaml`, true/false): Also give the definition in YAML, as Rundeck exports it. Default: `false`.

**Outputs**

- `average_duration_seconds` (number): How long a run takes on average, in seconds
- `definition` (object): Its whole definition, as Rundeck exports it in JSON
- `definition_yaml` (text): Its definition in YAML, as Rundeck exports it, when asked for; else empty
- `description` (text): Its description; empty if none
- `execution_enabled` (true/false): Whether it may run at all (by hand, by schedule or by the API)
- `full_name` (text): Its group and name, such as deploy/web/restart
- `group` (text): The group (folder) it's in, such as deploy/web; empty if none
- `id` (text): The job's ID (a UUID), as its page's address shows it
- `link` (text): A link to the job in Rundeck
- `log_level` (text): The log level it runs with, such as INFO
- `name` (text): The job's name
- `node_filter` (text): The nodes it runs on; empty when it runs on Rundeck's own
- `options` (list): What a run of it can be given, in order
- `project` (text): The project it's in
- `runs_locally` (true/false): Whether it runs on Rundeck's own server only
- `schedule` (text): Its schedule, as a Quartz cron line; empty if none
- `schedule_enabled` (true/false): Whether its schedule is on
- `scheduled` (true/false): Whether it has a schedule
- `step_count` (number): How many steps it has
- `timeout` (text): How long it may run before Rundeck stops it; empty if no limit

**Example output**

```json
{
  "average_duration_seconds": 11.8,
  "definition": {
    "group": "deploy/web",
    "id": "9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
    "name": "restart",
    "sequence": {
      "commands": [
        {
          "exec": "systemctl restart web"
        }
      ]
    }
  },
  "definition_yaml": "",
  "description": "Restart the web service",
  "execution_enabled": true,
  "full_name": "deploy/web/restart",
  "group": "deploy/web",
  "id": "9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
  "link": "https://rundeck.example.com:4440/project/ops/job/show/9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
  "log_level": "INFO",
  "name": "restart",
  "node_filter": "tags: web",
  "options": [
    {
      "allowed_values": [
        "staging",
        "production"
      ],
      "default": "staging",
      "description": "Where to restart it",
      "label": "Environment",
      "multivalued": false,
      "name": "environment",
      "required": true,
      "restricted": true,
      "secure": false
    }
  ],
  "project": "ops",
  "runs_locally": false,
  "schedule": "0 30 2 ? * * *",
  "schedule_enabled": true,
  "scheduled": true,
  "step_count": 1,
  "timeout": "10m"
}
```

### Get project

Get a project's name, description and link, with how many jobs it has and how many are scheduled.

**Inputs**

- Project (`project`, text, required): The project, by its name as Rundeck lists it, such as ops.

**Outputs**

- `created` (text): When it was made (ISO 8601); empty if Rundeck doesn't say
- `description` (text): Its description; empty if none
- `job_count` (number): How many jobs it has
- `label` (text): The name Rundeck shows for it; its name if it has no label
- `link` (text): A link to the project in Rundeck
- `name` (text): The project's name, such as ops
- `scheduled_job_count` (number): How many of them have a schedule

**Example output**

```json
{
  "created": "2026-01-05T09:12:00Z",
  "description": "Runbooks for the web tier",
  "job_count": 12,
  "label": "Operations",
  "link": "https://rundeck.example.com:4440/project/ops/home",
  "name": "ops",
  "scheduled_job_count": 3
}
```

### Get system info

Get the Rundeck server's version, whether it's running jobs (active) or not (passive), how many are running and how long it has been up.

**Inputs**: None.

**Outputs**

- `api_version` (number): The newest API version it answers
- `build` (text): Its build
- `execution_mode` (text): active or passive
- `executions_active` (true/false): Whether it runs jobs (active); false when executions are turned off (passive)
- `java_version` (text): The Java it runs on
- `link` (text): A link to its System Report page
- `memory_used_percent` (number): How much of its memory it uses, in percent
- `os` (text): Its operating system
- `running_jobs` (number): How many jobs are running now
- `server_name` (text): The server's name
- `server_uuid` (text): The server's ID, which tells cluster members apart
- `up_since` (text): When it started (ISO 8601)
- `uptime_seconds` (number): How long it has been up, in seconds
- `version` (text): Rundeck's version, such as 5.17.0-20251111

**Example output**

```json
{
  "api_version": 56,
  "build": "5.17.0-20251111",
  "execution_mode": "active",
  "executions_active": true,
  "java_version": "17.0.12",
  "link": "https://rundeck.example.com:4440/menu/systemInfo",
  "memory_used_percent": 41.5,
  "os": "Linux 6.1 amd64",
  "running_jobs": 2,
  "server_name": "rundeck1",
  "server_uuid": "a14bc3e6-75e8-4fe4-a90d-a16dcc976bf6",
  "up_since": "2026-03-01T10:15:00Z",
  "uptime_seconds": 86400.0,
  "version": "5.17.0-20251111"
}
```

### Import job

Make or update a job in a project from its definition in YAML or JSON, as Rundeck exports it (Job Actions > Download) and 'Get job' gives it.

**Inputs**

- Project (`project`, text, required): The project, by its name as Rundeck lists it, such as ops.
- Definition (`definition`, text, required): The job's definition, YAML or JSON. Several jobs in one definition are all imported.
- If the job exists (`if_exists`, text): What to do when the project has a job with the same ID, or the same group and name. One of: `update` (Update it), `skip` (Leave it as it is), `create` (Make another job with the same name). Default: `update`.
- Keep its ID (`keep_id`, true/false): Keep the ID in the definition. Turn off to copy a job, so the copy gets an ID of its own. Default: `true`.

**Outputs**

- `failed` (number): How many Rundeck refused
- `imported` (number): How many jobs were made or updated
- `job_id` (text): The ID of the (first) job imported or skipped
- `jobs` (list): Each job in the definition and what happened to it
- `project` (text): The project
- `skipped` (number): How many were left as they were

**Example output**

```json
{
  "failed": 0,
  "imported": 1,
  "job_id": "9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
  "jobs": [
    {
      "error": "",
      "full_name": "deploy/web/restart",
      "group": "deploy/web",
      "id": "9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
      "link": "https://rundeck.example.com:4440/project/ops/job/show/9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
      "name": "restart",
      "status": "succeeded"
    }
  ],
  "project": "ops",
  "skipped": 0
}
```

### List jobs

List or find a project's jobs by name and group, with whether each is scheduled and turned on, and its ID for the other job steps.

**Inputs**

- Project (`project`, text, required): The project, by its name as Rundeck lists it, such as ops.
- Name contains (`name`, text): Only jobs whose name has these letters, such as restart.
- Group (`group`, text): Only jobs in this group (folder) or a group inside it, such as deploy/web.
- Scheduled only (`scheduled_only`, true/false): Only jobs that have a schedule. Default: `false`.

**Outputs**

- `count` (number): How many there are
- `jobs` (list): The jobs, by group and name
- `project` (text): The project they're in

**Example output**

```json
{
  "count": 1,
  "jobs": [
    {
      "description": "Restart the web service",
      "execution_enabled": true,
      "full_name": "deploy/web/restart",
      "group": "deploy/web",
      "id": "9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
      "link": "https://rundeck.example.com:4440/project/ops/job/show/9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
      "name": "restart",
      "project": "ops",
      "schedule_enabled": true,
      "scheduled": true
    }
  ],
  "project": "ops"
}
```

### List keys

List what's in a Key Storage folder: its folders, and its passwords and keys by type and when they changed. Rundeck never gives a password or private key back.

**Inputs**

- Folder (`path`, text): The folder, as Key Storage shows it, such as keys/project/ops. Leave blank for the top.

**Outputs**

- `count` (number): How many there are
- `keys` (list): What's in it: folders first, then keys
- `path` (text): The folder listed

**Example output**

```json
{
  "count": 1,
  "keys": [
    {
      "created": "2026-02-01T08:00:00Z",
      "is_folder": false,
      "key_type": "password",
      "modified": "2026-02-20T08:00:00Z",
      "name": "db.password",
      "path": "keys/project/ops/db.password"
    }
  ],
  "path": "keys/project/ops"
}
```

### List nodes

List a project's nodes (the servers jobs and commands run on), all of them or those a node filter matches, with their host names, tags and attributes.

**Inputs**

- Project (`project`, text, required): The project, by its name as Rundeck lists it, such as ops.
- Node filter (`node_filter`, text): Only the nodes this filter matches. Rundeck's node filter, as its Nodes page takes it, such as name: web1, tags: web or .* for every node.

**Outputs**

- `count` (number): How many there are
- `nodes` (list): The nodes, by name
- `project` (text): The project

**Example output**

```json
{
  "count": 1,
  "nodes": [
    {
      "attributes": {
        "ssh-key-storage-path": "keys/project/ops/web.key"
      },
      "description": "Web server",
      "hostname": "web1.example.com",
      "name": "web1",
      "os_arch": "amd64",
      "os_family": "unix",
      "os_name": "Linux",
      "os_version": "6.1",
      "tags": [
        "prod",
        "web"
      ],
      "username": "rundeck"
    }
  ],
  "project": "ops"
}
```

### List projects

List the Rundeck projects the token may see, by name, with their descriptions.

**Inputs**: None.

**Outputs**

- `count` (number): How many there are
- `projects` (list): The projects, by name

**Example output**

```json
{
  "count": 1,
  "projects": [
    {
      "created": "2026-01-05T09:12:00Z",
      "description": "Runbooks for the web tier",
      "label": "Operations",
      "link": "https://rundeck.example.com:4440/project/ops/home",
      "name": "ops"
    }
  ]
}
```

### List running executions

List the jobs and commands running now, in one project or all of them, optionally of one job.

**Inputs**

- Project (`project`, text): Only executions in this project. Leave blank for every project.
- Job (`job_id`, text): Only executions of this job.

**Outputs**

- `count` (number): How many there are
- `executions` (list): The executions, the latest first
- `project` (text): The project, or * for every project

**Example output**

```json
{
  "count": 1,
  "executions": [
    {
      "argstring": "-environment staging",
      "custom_status": "",
      "description": "systemctl restart web",
      "duration_seconds": 12.4,
      "ended": "",
      "execution_type": "user",
      "failed_nodes": [],
      "finished": false,
      "id": 123,
      "job_full_name": "deploy/web/restart",
      "job_group": "deploy/web",
      "job_id": "9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
      "job_name": "restart",
      "link": "https://rundeck.example.com:4440/project/ops/execution/show/123",
      "options": {
        "environment": "staging"
      },
      "project": "ops",
      "started": "2026-03-02T10:15:00Z",
      "status": "running",
      "succeeded": false,
      "successful_nodes": [],
      "user": "lodol"
    }
  ],
  "project": "ops"
}
```

### Run command

Run a command on the nodes you choose, as Rundeck's Commands page does, such as uptime to check servers. Turn on waiting to get its output.

**Inputs**

- Project (`project`, text, required): The project, by its name as Rundeck lists it, such as ops.
- Command (`command`, text, required): The command to run on each node, such as uptime or df -h.
- Nodes (`node_filter`, text, required): Which nodes to run it on. Rundeck's node filter, as its Nodes page takes it, such as name: web1, tags: web or .* for every node.
- Nodes at once (`thread_count`, number): How many nodes to run it on at the same time, 1 to 100. Defaults to 1, one after another. From 1 to 100.
- Keep going on failure (`keep_going`, true/false): Carry on with the other nodes when it fails on one. Default: `false`.
- Wait for it to finish (`wait`, true/false): Wait until the command stops, then return how it went and the end of its log. Off returns as soon as it starts. Default: `false`.
- Wait up to (minutes) (`max_wait_minutes`, number): How long to wait, 1 to 15 minutes. Defaults to 10. It keeps running in Rundeck after. Only when Wait for it to finish is `true`. From 1 to 15.

**Outputs**

- `argstring` (text): The options it ran with, as Rundeck writes them
- `custom_status` (text): The status a job set itself, when status is other
- `description` (text): What it ran: the job's steps, or the command
- `duration_seconds` (number): How long it ran, or has run so far, in seconds
- `ended` (text): When it stopped (ISO 8601); empty while it runs
- `execution_type` (text): user (by a person or the API), scheduled or user-scheduled
- `failed_nodes` (list): The nodes it failed on
- `finished` (true/false): Whether it has stopped
- `id` (number): The execution's number, as Rundeck shows it (#123)
- `job_full_name` (text): The job's group and name; empty for a command
- `job_group` (text): The job's group; empty if none
- `job_id` (text): The job's ID; empty for a command
- `job_name` (text): The job's name; empty for a command
- `link` (text): A link to the execution in Rundeck
- `options` (object): The options it ran with, by name
- `output` (text): The end of its log, once waited for; empty otherwise
- `project` (text): The project it ran in
- `started` (text): When it started (ISO 8601); empty until it does
- `status` (text): running, succeeded, failed, aborted, timedout, failed-with-retry, scheduled (set to run later) or other (a status the job set itself)
- `succeeded` (true/false): Whether it succeeded
- `successful_nodes` (list): The nodes it succeeded on
- `timed_out` (true/false): Whether it was still running when the wait was up
- `user` (text): Who started it

**Example output**

```json
{
  "argstring": "",
  "custom_status": "",
  "description": "uptime",
  "duration_seconds": 12.4,
  "ended": "2026-03-02T10:15:12Z",
  "execution_type": "user",
  "failed_nodes": [],
  "finished": true,
  "id": 123,
  "job_full_name": "",
  "job_group": "",
  "job_id": "",
  "job_name": "",
  "link": "https://rundeck.example.com:4440/project/ops/execution/show/123",
  "options": {},
  "output": "10:15:01 up 12 days,  2 users,  load average: 0.10, 0.08, 0.05",
  "project": "ops",
  "started": "2026-03-02T10:15:00Z",
  "status": "succeeded",
  "succeeded": true,
  "successful_nodes": [
    "web1"
  ],
  "timed_out": false,
  "user": "lodol"
}
```

### Run job

Run a job with the options you give, as Rundeck's Run Job Now button does, on its nodes or the ones you choose, now or at a set time. Turn on waiting to get how it went and the end of its log.

**Inputs**

- Job (`job_id`, text, required): The job: pick one, or paste its ID or a link to its page in Rundeck.
- Options (`options`, object): The job's options by name, such as {"environment": "staging"}. Leave blank for the defaults; "Get job" lists them.
- Nodes (`node_filter`, text): Run it on these nodes instead of the job's own. Rundeck's node filter, as its Nodes page takes it, such as name: web1, tags: web or .* for every node.
- Log level (`log_level`, text): How much it logs. Debug shows more; Normal is the job's usual. One of: `INFO` (Normal), `VERBOSE`, `DEBUG`, `WARN` (Warnings and errors only), `ERROR` (Errors only). Default: `INFO`.
- Run at (`run_at`, date and time): When to run it. Leave blank to run it now.
- Wait for it to finish (`wait`, true/false): Wait until the job stops, then return how it went and the end of its log. Off returns as soon as it starts. Default: `false`.
- Wait up to (minutes) (`max_wait_minutes`, number): How long to wait, 1 to 15 minutes. Defaults to 10. It keeps running in Rundeck after. Only when Wait for it to finish is `true`. From 1 to 15.

**Outputs**

- `argstring` (text): The options it ran with, as Rundeck writes them
- `custom_status` (text): The status a job set itself, when status is other
- `description` (text): What it ran: the job's steps, or the command
- `duration_seconds` (number): How long it ran, or has run so far, in seconds
- `ended` (text): When it stopped (ISO 8601); empty while it runs
- `execution_type` (text): user (by a person or the API), scheduled or user-scheduled
- `failed_nodes` (list): The nodes it failed on
- `finished` (true/false): Whether it has stopped
- `id` (number): The execution's number, as Rundeck shows it (#123)
- `job_full_name` (text): The job's group and name; empty for a command
- `job_group` (text): The job's group; empty if none
- `job_id` (text): The job's ID; empty for a command
- `job_name` (text): The job's name; empty for a command
- `link` (text): A link to the execution in Rundeck
- `options` (object): The options it ran with, by name
- `output` (text): The end of its log, once waited for; empty otherwise
- `project` (text): The project it ran in
- `started` (text): When it started (ISO 8601); empty until it does
- `status` (text): running, succeeded, failed, aborted, timedout, failed-with-retry, scheduled (set to run later) or other (a status the job set itself)
- `succeeded` (true/false): Whether it succeeded
- `successful_nodes` (list): The nodes it succeeded on
- `timed_out` (true/false): Whether it was still running when the wait was up
- `user` (text): Who started it

**Example output**

```json
{
  "argstring": "-environment staging",
  "custom_status": "",
  "description": "systemctl restart web",
  "duration_seconds": 12.4,
  "ended": "2026-03-02T10:15:12Z",
  "execution_type": "user",
  "failed_nodes": [],
  "finished": true,
  "id": 123,
  "job_full_name": "deploy/web/restart",
  "job_group": "deploy/web",
  "job_id": "9e5c1a2b-1c2d-4e5f-8a9b-0c1d2e3f4a5b",
  "job_name": "restart",
  "link": "https://rundeck.example.com:4440/project/ops/execution/show/123",
  "options": {
    "environment": "staging"
  },
  "output": "Restarting web\nweb is running",
  "project": "ops",
  "started": "2026-03-02T10:15:00Z",
  "status": "succeeded",
  "succeeded": true,
  "successful_nodes": [
    "web1"
  ],
  "timed_out": false,
  "user": "lodol"
}
```

### Store key

Add a password, private key or public key to Key Storage, or replace the one at that path, such as after rotating a database password jobs use.

**Inputs**

- Path (`path`, text, required): Where to keep it, folders and a name, such as keys/project/ops/db.password.
- Type (`key_type`, text): What it is. One of: `password`, `private` (Private key), `public` (Public key). Default: `password`.
- Value (`value`, text, required): The password, or the whole key as text (a private key starts with -----BEGIN).

**Outputs**

- `created` (true/false): Whether it was added; false if one was there and was replaced
- `key_type` (text): password, private or public
- `modified` (text): When it was stored (ISO 8601)
- `name` (text): Its name
- `path` (text): The key's path

**Example output**

```json
{
  "created": false,
  "key_type": "password",
  "modified": "2026-03-02T10:15:00Z",
  "name": "db.password",
  "path": "keys/project/ops/db.password"
}
```
