> ## 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.

# Cluster Setup

In the directory where the `config.json` file resides, you will find a script named `setup_cluster.sh`, which performs the automatic deployment of Regatta.

Switch to the `deployment_user` and run `setup_cluster.sh` (RDT) to begin the automated deployment:

```shell theme={null}
./setup_cluster.sh
```

You will be asked to accept the Regatta End-User License Agreement (EULA), located at: [https://regatta.dev/eula/](https://regatta.dev/eula/)

The prompt will appear as follows:

```shell theme={null}
By using this script and deploying the Regatta Cluster you agree to the EULA located at:
https://regatta.dev/eula/
Please read the EULA and accept the terms.
Do you accept the terms of the EULA? [Y/n]
```

If you need time to review the EULA or choose not to continue, entering `n` will abort the script:

```shell theme={null}
EULA not accepted - deployment aborted...
```

You may restart the deployment at any time using the same script.

By entering `Y`, you confirm that you have read and accepted the EULA and the deployment will continue.

Next, the script will prompt you to provide the path to the SSH key and the username used to connect to the cluster nodes:

```shell theme={null}
Enter path to SSH key: /<path>/<ssh_key_file>
```

```shell theme={null}
Enter username to use with ssh key: deployment_user
```

<Info>
  SSH key is required by Ansible for connecting to the cluster nodes and perform the deployment.

  In case of a single-node deployment in which the RDT runs on the same node used for Regatta DB, you can create an SSH key-pair using the following command:

  ```bash theme={null}
  ssh-keygen -t ed25519 -f ~/.ssh/id_rdt -N "" && cat ~/.ssh/id_rdt.pub >> ~/.ssh/authorized_keys
  ```
</Info>

The RDT will then validate SSH connectivity to every node defined in its configuration file. After successful validation, it will ask whether the `deployment_user` requires a password for sudo operations:

```shell theme={null}
Does the <deployment_user> require a password to run sudo commands? [Y/n]:
```

If the answer is yes, you will be prompted to enter the password later:

```shell theme={null}
BECOME PASS:
```

<Note>
  **The password remains only in volatile memory and is never stored.**
</Note>

<Warning>
  During execution, the RDT installs the supplied RPM packages. These packages check for the Regatta dependencies listed in the [prerequisites](/self-hosted-deployment/prerequisites) section. If dependencies are missing and no internet connection is available to download them, the deployment will stop at this stage.
</Warning>

The RDT will then run the deployment sequence, performing the following tasks:

1. Read the `config.json` file and parse the list of IP addresses and storage device paths
2. Connect to each node using the configured `deployment_user`
3. Copy and install the Regatta RPMs on each node
4. Change ownership of the block storage device(s) to the `regatta` user
5. Generate the Regatta `cluster.config` file and distribute it to all nodes
6. Start the Regatta services, which in turn start all Regatta modules

Once the RDT completes its execution, the Regatta cluster will be fully operational and ready to accept connections via the CLI or client drivers.

You can verify that the cluster is running by checking the Regatta services on any node:

```shell theme={null}
sudo systemctl status regatta-1.0-*
```

Example outuput (each Regatta module runs under its own service, such as `regatta-1.0-rdb.service`):

```shell theme={null}
● regatta-1.0-gdd.service - The Regatta gdd service
   Loaded: loaded (/usr/lib/systemd/system/regatta-1.0-gdd.service; enabled; vendor preset: disabled)
   Active: active (running) since Sun 2025-11-23 18:00:16 IST; 22h ago
 Main PID: 3331 (start.sh)
    Tasks: 18 (limit: 203820)
   Memory: 143.6M
   CGroup: /system.slice/regatta-1.0-gdd.service
           ├─3331 /bin/sh /opt/regatta/1.0/gdd/bin/start.sh gdd
           └─3384 /opt/regatta/1.0/gdd/bin/rdb.bin --node 2 --role gdd --sys_conf /opt/regatta/1.0/gdd/conf/cluster.config --core_conf /opt/regatta/1.0/gdd/conf/cluster.config --logs_dir /var/log/regatta/g>

Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3331]: Starting module gdd with number '2' and using path /opt/regatta/1.0/gdd
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3384]: Setting module ID to 2
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3384]: Module PID is 3384
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3384]: Setting node role to gdd
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3384]: Setting non-default configuration file /opt/regatta/1.0/gdd/conf/cluster.config
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3384]: Setting core configuration file /opt/regatta/1.0/gdd/conf/cluster.config
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3384]: Setting logs directory path /var/log/regatta/gdd
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3384]: Setting node boot mode to NonGenesis
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3384]: Starting the database module
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3384]: Core loaded

● regatta-1.0-dcm.service - The Regatta dcm service
   Loaded: loaded (/usr/lib/systemd/system/regatta-1.0-dcm.service; enabled; vendor preset: disabled)
   Active: active (running) since Sun 2025-11-23 18:00:17 IST; 22h ago
 Main PID: 3417 (start.sh)
    Tasks: 18 (limit: 203820)
   Memory: 202.5M
   CGroup: /system.slice/regatta-1.0-dcm.service
           ├─3417 /bin/sh /opt/regatta/1.0/dcm/bin/start.sh dcm
           └─3429 /opt/regatta/1.0/dcm/bin/rdb.bin --node 3 --role dcm --sys_conf /opt/regatta/1.0/dcm/conf/cluster.config --core_conf /opt/regatta/1.0/dcm/conf/cluster.config --logs_dir /var/log/regatta/d>

Nov 23 18:00:17 rgtv-8c257800-000-sgdcpr start.sh[3417]: Starting module dcm with number '3' and using path /opt/regatta/1.0/dcm
Nov 23 18:00:17 rgtv-8c257800-000-sgdcpr start.sh[3429]: Setting module ID to 3
Nov 23 18:00:17 rgtv-8c257800-000-sgdcpr start.sh[3429]: Module PID is 3429
Nov 23 18:00:17 rgtv-8c257800-000-sgdcpr start.sh[3429]: Setting node role to dcm
Nov 23 18:00:17 rgtv-8c257800-000-sgdcpr start.sh[3429]: Setting non-default configuration file /opt/regatta/1.0/dcm/conf/cluster.config
Nov 23 18:00:17 rgtv-8c257800-000-sgdcpr start.sh[3429]: Setting core configuration file /opt/regatta/1.0/dcm/conf/cluster.config
Nov 23 18:00:17 rgtv-8c257800-000-sgdcpr start.sh[3429]: Setting logs directory path /var/log/regatta/dcm
Nov 23 18:00:17 rgtv-8c257800-000-sgdcpr start.sh[3429]: Setting node boot mode to Genesis
Nov 23 18:00:17 rgtv-8c257800-000-sgdcpr start.sh[3429]: Starting the database module
Nov 23 18:00:17 rgtv-8c257800-000-sgdcpr start.sh[3429]: Core loaded

● regatta-1.0-rdb.service - The Regatta rdb service
   Loaded: loaded (/usr/lib/systemd/system/regatta-1.0-rdb.service; enabled; vendor preset: disabled)
   Active: active (running) since Sun 2025-11-23 18:00:19 IST; 22h ago
 Main PID: 3497 (start.sh)
    Tasks: 22 (limit: 203820)
   Memory: 1.9G
   CGroup: /system.slice/regatta-1.0-rdb.service
           ├─3497 /bin/sh /opt/regatta/1.0/rdb/bin/start.sh rdb
           └─3509 /opt/regatta/1.0/rdb/bin/rdb.bin --node 10 --role rdb --sys_conf /opt/regatta/1.0/rdb/conf/cluster.config --core_conf /opt/regatta/1.0/rdb/conf/cluster.config --logs_dir /var/log/regatta/>

Nov 23 18:00:19 rgtv-8c257800-000-sgdcpr start.sh[3497]: Starting module rdb with number '10' and using path /opt/regatta/1.0/rdb
Nov 23 18:00:19 rgtv-8c257800-000-sgdcpr start.sh[3509]: Setting module ID to 10
Nov 23 18:00:19 rgtv-8c257800-000-sgdcpr start.sh[3509]: Module PID is 3509
Nov 23 18:00:19 rgtv-8c257800-000-sgdcpr start.sh[3509]: Setting node role to rdb
Nov 23 18:00:19 rgtv-8c257800-000-sgdcpr start.sh[3509]: Setting non-default configuration file /opt/regatta/1.0/rdb/conf/cluster.config
Nov 23 18:00:19 rgtv-8c257800-000-sgdcpr start.sh[3509]: Setting core configuration file /opt/regatta/1.0/rdb/conf/cluster.config
Nov 23 18:00:19 rgtv-8c257800-000-sgdcpr start.sh[3509]: Setting logs directory path /var/log/regatta/rdb
Nov 23 18:00:19 rgtv-8c257800-000-sgdcpr start.sh[3509]: Setting node boot mode to Genesis
Nov 23 18:00:19 rgtv-8c257800-000-sgdcpr start.sh[3509]: Starting the database module
Nov 23 18:00:19 rgtv-8c257800-000-sgdcpr start.sh[3509]: Core loaded

● regatta-1.0-sequencer.service - The Regatta sequencer service
   Loaded: loaded (/usr/lib/systemd/system/regatta-1.0-sequencer.service; enabled; vendor preset: disabled)
   Active: active (running) since Sun 2025-11-23 18:00:16 IST; 22h ago
 Main PID: 3369 (start.sh)
    Tasks: 18 (limit: 203820)
   Memory: 123.8M
   CGroup: /system.slice/regatta-1.0-sequencer.service
           ├─3369 /bin/sh /opt/regatta/1.0/sequencer/bin/start.sh sequencer
           └─3385 /opt/regatta/1.0/sequencer/bin/rdb.bin --node 1 --role sequencer --sys_conf /opt/regatta/1.0/sequencer/conf/cluster.config --core_conf /opt/regatta/1.0/sequencer/conf/cluster.config --log>

Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3369]: Starting module sequencer with number '1' and using path /opt/regatta/1.0/sequencer
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3385]: Setting module ID to 1
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3385]: Module PID is 3385
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3385]: Setting node role to sequencer
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3385]: Setting non-default configuration file /opt/regatta/1.0/sequencer/conf/cluster.config
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3385]: Setting core configuration file /opt/regatta/1.0/sequencer/conf/cluster.config
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3385]: Setting logs directory path /var/log/regatta/sequencer
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3385]: Setting node boot mode to Genesis
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3385]: Starting the database module
Nov 23 18:00:16 rgtv-8c257800-000-sgdcpr start.sh[3385]: Core loaded
```

Congratulations! Your Regatta cluster is up and running. Now you can connect your application using your preferred [client](/drivers-and-clients/connecting-applications-to-regatta).