> ## Documentation Index
> Fetch the complete documentation index at: https://docs.regatta.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Quickstart: Single-Node RegattaDB

> Install RDT and stand up a single-node RegattaDB deployment end to end.

This page walks you through creating a single-node RegattaDB deployment from scratch. Make sure you've read
[Prerequisites](../prerequisites.mdx) first: you'll need SSH access (key or password)
from your control node to the node you're deploying to.

## 1. Install RDT

```bash theme={null}
sudo dnf -y install regatta-rdt-<version>.el8.x86_64.rpm
```

See [Installation](../installation.mdx) for other install paths.

## 2. Set up SSH access

RDT connects to your node over SSH using either a key or a password. If your SSH user,
key, or password aren't already set up as RDT's defaults, pass them explicitly on every
command:

```bash theme={null}
rdt --user regatta --key ~/.ssh/id_rsa <command> ...
```

See [Global Options & Credentials](../advanced/global-options-and-credentials.mdx) for
all connection options; the rest of this walkthrough omits them for brevity.

## 3. Create a config

```bash theme={null}
rdt config create --name quickstart --from xsmall --ip-list 10.0.0.1
```

This uses the `xsmall` template: a single node running every module type. Replace
`10.0.0.1` with your node's real address.

The template doesn't specify storage devices, and every RDB module needs at least one.
Edit the config and add one before deploying:

```bash theme={null}
rdt config edit quickstart
```

```json theme={null}
"devices": [
  { "name": "disk1", "path": "/dev/nvme0n1" }
],
"modules": [
  { "type": "rdb", "id": 11, "devices": ["disk1"], ... }
]
```

Everything else (CPU thread count, RAM) can be left unset, the System Manager decides
those automatically, and storage devices can be auto-detected instead of hand-edited;
see the next step.

Review the result:

```bash theme={null}
rdt config print quickstart --validate
```

See [Configuration Overview](../configuration/overview.mdx) for the full schema.

## 4. Probe node hardware (optional)

Instead of hand-editing CPU/RAM/device info, you can have RDT connect to the node and
detect it for you when creating the config:

```bash theme={null}
rdt config create --name quickstart --from xsmall --ip-list 10.0.0.1 --probe-nodes
```

RDT looks for unmounted, unformatted block devices on the node and assigns any it finds
(above \~20GB) to the RDB module automatically. If the node has no such device attached,
you'll still need to add one manually as shown in step 3.

## 5. Create and start RegattaDB

```bash theme={null}
rdt create-cluster --name quickstart --src /path/to/regatta-rpms
```

`--src` points to a directory containing the Regatta RPMs (a `.tar.zst` archive also
works; see [RegattaDB Lifecycle](../system-setup/lifecycle.mdx)).

If something goes wrong partway through, see
[RegattaDB Lifecycle](../system-setup/lifecycle.mdx) for how to retry individual
steps, and [Collecting Debug Information](./collecting-debug-info.mdx) if you need to
gather logs.

## 6. Verify and connect

Once `create-cluster` finishes, connect to RegattaDB to confirm it's up and start
running SQL.

### Connection parameters

* **Username**: `admin`
* **Password**: `RegattaDefault1234!` (the default admin credentials; change this
  immediately after deployment)
* **URL**: `10.0.0.1:8840` (your node's IP and the SM's default communication port)

### Connect with the RegattaDB CLI

```bash theme={null}
/opt/regatta/<major.minor>/clients/cli/bin/client_cli
```

```
rdb>\connect user=admin pass=RegattaDefault1234! url=10.0.0.1:8840
```

If the connection succeeds:

```
Successfully connected to RegattaDB system
```

See [Connecting Applications to Regatta](/drivers-and-clients/connecting-applications-to-regatta)
to connect from your own application instead.

### Confirm the system is active

```sql theme={null}
regatta > MANAGE SYSTEM SHOW;
statement executed successfully
| STATE  |
----------
| Active |
```

A healthy single-node deployment reports `Active`. See
[System States and Commands](https://docs.regatta.dev/sql/system-management/system-states-and-commands#system-show)
for what each state means.

## Next steps

* [Multi-Node Deployment](./multi-node-deployment.mdx) for a realistic production topology.
* [Config Recipes](./config-recipes.mdx) for hardware probing and other config shortcuts.

## Appendix: full config file

```json expandable theme={null}
{
  "version": "26.0",
  "id": "quickstart",
  "nodes": [
    {
      "id": 1,
      "name": "node_1",
      "ip": "10.0.0.1",
      "devices": [
        { "name": "disk1", "path": "/dev/nvme0n1" }
      ],
      "modules": [
        { "type": "sna", "id": 10, "port": 8841, "service_port": 5001, "config": {}, "devices": [] },
        { "type": "rdb", "id": 11, "port": 8850, "service_port": 6001, "config": {}, "devices": ["disk1"] },
        { "type": "sm", "id": 1, "port": 8840, "service_port": 5000, "config": {}, "devices": [] },
        { "type": "sequencer", "id": 3, "port": 8842, "service_port": 5002, "config": {}, "devices": [] },
        { "type": "gdd", "id": 4, "port": 8843, "service_port": 5003, "config": {}, "devices": [] },
        { "type": "dcm", "id": 5, "port": 8844, "service_port": 5004, "config": {}, "devices": [] }
      ]
    }
  ]
}
```