Deploy with Launchpad

Launchpad is a command-line interface (CLI) tool that provides a quick, comprehensive solution to deploying Rocket.Chat with Kubernetes. It simplifies the deployment process by configuring the Kubernetes resources and supporting infrastructure required to run a Rocket.Chat workspace. Launchpad supports EKS and K3s.

Key benefits

• Automated orchestration: Handles end-to-end workspace deployments on Kubernetes clusters without manual setup of ingress controllers or TLS cert managers.

• Built-in observability: Comes pre-configured with Prometheus Operator and Grafana dashboards for immediate insight into workspace and infrastructure metrics.

• Streamlined database lifecycle: Supports either a fully managed built-in MongoDB operator or seamless integration with an external MongoDB cluster (such as MongoDB Atlas).

• Scaling: Scales the Rocket.Chat application services from the CLI as your usage changes.

Launchpad is intended for users who want a guided deployment experience without manually assembling every component of the Kubernetes stack. If you need more control over the individual components, see the Kubernetes, Docker, or Podman deployment guides.

This guide walks you through deploying Rocket.Chat with Launchpad, from preparing your environment to accessing and monitoring your workspace.

Before you begin, review the following limitations:

• Launchpad currently does not support Omnichannel or federation workloads. For these use cases, use one of the other recommended deployment methods.

• It is recommended that you first deploy in a test environment to verify that the workspace is deployed according to your needs before deploying to production.

Launchpad architecture components

Here is an overview of the internal components that Launchpad uses to deploy workspaces:

Launchpad workspace sizing guide

Launchpad sizes your workspace using a preset. Each preset defines the number of replicas, MongoDB resources and storage, and minimum cluster capacity.  

Choose a preset that aligns with your team before deploying:

These figures are sizing guidance rather than hard limits. Actual capacity depends on factors such as message volume, channel size, WebSocket connections, and usage patterns.

Each preset deploys the following number of replicas for Rocket.Chat microservice:

You can change these counts after deployment, as described in Scale your workspace.

Prerequisites

This guide primarily uses AWS EKS to deploy Rocket.Chat with Launchpad using the small preset.

Before installing Launchpad, make sure you have the following:

1. An AWS account with appropriate permissions and credentials.

2. AWS CLI installed and configured.

   1. Follow the AWS CLI installation instructions for your operating system.

   2. After installing the AWS CLI, configure your AWS credentials according to your organization requirements. For example, you can use the aws configure command, environment variables, or a config file. See the AWS CLI Getting Started section for details.

3. eksctl installed.

4. kubectl installed.

5. A custom domain for the workspace with access to your DNS provider.

6. File storage sytem for storing workspace files. Amazon S3 is recommended.

If you are using K3s instead of EKS, you need:

• A ready K3s cluster with Traefik disabled (--disable=traefik).

• Longhorn dependencies installed to provide storage volumes managed by Launchpad.

Step 1: Choose a database strategy

You can either use the MongoDB deployment that ships with Launchpad or connect your own external database. This choice affects the size of your cluster.

• Built-in MongoDB: Best for deployments where you do not want to manage a separate database.

• External MongoDB: Deployments where you already run MongoDB. MongoDB Atlas is recommended.

Cluster sizing guide

The following table shows the minimum cluster requirements for each Launchpad preset. External MongoDB deployments require fewer cluster nodes because MongoDB runs outside the Kubernetes cluster.

For MongoDB Atlas, it’s recommended to host the cluster in the same AWS region as your deployment, or as close to it as possible, and use peering to connect it to your AWS environment.

You’ll need the MongoDB Atlas connection string during deployment. It should follow this format:

mongodb+srv://<user>:<pass>@cluster1.q9p55.mongodb.net/db

Step 2: Install Launchpad CLI

1. Download the Launchpad archive matching your operating system and CPU architecture:

   1. Linux (amd64) - https://go.rocket.chat/i/launchpad-linux-amd64-latest

   2. Linux (arm) - https://go.rocket.chat/i/launchpad-linux-arm-latest

   3. Linux (arm64) - https://go.rocket.chat/i/launchpad-linux-arm64-latest

   4. macOS (amd64) - https://go.rocket.chat/i/launchpad-darwin-amd64-latest

   5. macOS (arm64) - https://go.rocket.chat/i/launchpad-darwin-arm64-latest

   6. Windows (amd64) - https://go.rocket.chat/i/launchpad-windows-amd64-latest

   7. Checksums (latest) - https://go.rocket.chat/i/launchpad_checksums-latest

2. Extract the downloaded archive.

   The extracted directory contains the Launchpad binary and an example deployment configuration:

   launchpad
   examples/
   └── deploy.yaml

3. From the directory containing the Launchpad binary, run:

   ./launchpad version

   You should see output similar to:

   launchpad version 0.4.0, blueprints v1.2.0-alpha1, commit 4cf34edcaffd030ab00da08834eca9ea1abc3036 (2026-01-29T20:19:20Z), go version go1.24.4

   You can also run ./launchpad to display the available commands.

If the command is not found, make sure you are running it from the directory containing the launchpad binary.

Step 3: Create the Kubernetes cluster

Before you continue, confirm that you have configured your AWS credentials on the AWS CLI. If you haven’t, see the AWS CLI Getting Started section for details.

1. In the directory that contains the Launchpad binary, create a file named eks-basic.yaml .

2. Add the following configuration to the file based on your database strategy:

# eks-basic.yaml
# More examples at https://github.com/eksctl-io/eksctl/tree/main/examples
apiVersion: eksctl.io/v1alpha5
kind: ClusterConfig
metadata:
  name: dev-eks-launchpad # This is the cluster name. You can change the name if needed.
  region: us-east-1 # Change the client region if it is other than us-east-1.
  version: "1.33" # Kubernetes version. Change the value if required.
addons:
  - name: aws-ebs-csi-driver # required for monitoring, MongoDB
    version: 1.47.0
vpc:
  cidr: 10.0.0.0/16
nodeGroups:
  - name: dev-eks-node-group
    instanceType: t3.large  # To deploy with built-in MongoDB, at least t3.large is required.
    desiredCapacity: 5      # To deploy built-in MongoDB, at least 5 is required.
    privateNetworking: true
    iam:
      withAddonPolicies:
        ebs: true     

# eks-basic.yaml
# More examples at https://github.com/eksctl-io/eksctl/tree/main/examples
apiVersion: eksctl.io/v1alpha5
kind: ClusterConfig
metadata:
  name: dev-eks-launchpad # This is the cluster name. You can change the name if you need to.
  region: us-east-1 # Change the client region if it is other than us-east-1.
  version: "1.33" # Kubernetes version. Change the value if required.
addons:
  - name: aws-ebs-csi-driver # required for monitoring, MongoDB
    version: 1.47.0
vpc:
  cidr: 10.0.0.0/16
nodeGroups:
  - name: dev-eks-node-group
    instanceType: t3.large  # At least t3.large is required.
    desiredCapacity: 3       # At least 3 is required.
    privateNetworking: true
    iam:
      withAddonPolicies:
        ebs: true            # required for monitoring, MongoDB

The cluster configuration above is for the small preset.

3. Save the file and create the cluster:

   eksctl create cluster -f eks-basic.yaml --kubeconfig=basic.config

   If your AWS account setup requires a named profile, use this command instead:

   eksctl --profile <aws-account-profile-name> create cluster -f eks-basic.yaml --kubeconfig=basic.config

   Cluster creation takes approximately 15–20 minutes. The command also creates a basic.config kubeconfig file containing the connection information for the new cluster.

4. After the cluster is created, verify that the nodes are available:

   kubectl get nodes

   You should see five nodes for a built-in MongoDB deployment, or three for an external database.
   

Step 4: Configure your deployment

Launchpad provides two ways to configure your deployment:

• Interactive terminal UI: Use Launchpad's guided terminal interface to create your deployment configuration. The interactive wizard prompts you for configuration details, validates your inputs, and can either deploy immediately or export a complete configuration file for future use. Use this if you are deploying for the first time or want to see every available option.

• Manual configuration: Edit the deploy.yaml file in the /examples folder and deploy it. Use this for repeatable setups, or when you already know the values you need.

Both methods produce the configuration file, so you can start in the wizard, export the file, and edit it later.

mongodb+srv://<user>:<password>@cluster.example.mongodb.net/rocketchat

./launchpad package

Step 5: Deploy the workspace

From the Launchpad binary directory, run the following to deploy the workspace to your Kubernetes cluster:

./launchpad deploy -f <config-file> -k basic.config -p eks

If you are using K3s, deploy with the -p k3s flag as follows:

./launchpad deploy -f <config-file> -k basic.config -p k3s

Replace <config-file> with the path to your configuration YAML file.

Configure your DNS record

On the first deployment, Launchpad performs a series of configuration checks. If the configuration is valid, the deployment pauses and Launchpad displays a CNAME target address. You must add this CNAME record to your DNS provider before continuing with the deployment.

The following screenshot shows an example of the CNAME target displayed by Launchpad:

1. Copy the CNAME target displayed by Launchpad.

2. Add the CNAME record to your DNS provider. Map your workspace hostname to the CNAME target provided by Launchpad.

3. Wait for the DNS record to propagate. This usually takes 1–2 minutes.

4. Run the same deployment command again.

After you rerun the deployment command, Launchpad detects the DNS record and continues the deployment. It completes the Let's Encrypt ACME HTTP challenge, provisions the TLS certificate, and initializes the workspace.

The following screenshot shows an example of a successful deployment:

The workspace is now available at your configured URL within a few minutes.

Step 6: Access your workspace

Great! You’ve successfully created your Rocket.Chat workspace with Launchpad. Access it using the workspace URL that you have defined in the configuration.

After the workspace loads, follow the on-screen prompts to create your first admin user and complete the initial workspace setup. Your workspace and email will be registered with the Rocket.Chat Cloud portal during this process.

Next, explore the following resources to continue using your new workspace:

• User Guides: Learn the basics of your Rocket.Chat account, the types of rooms, and how to communicate with your workspace users.

• Workspace Administration: Administrators and owners can set and manage various configurations. To share files on the workspace, make sure to configure the workspace file upload settings.

• Marketplace: Explore the available apps to enhance your workspace.

Monitor your workspace

Launchpad provides a Grafana dashboard for monitoring workspace metrics and logs. The dashboard is available through a local port-forwarding connection.

1. From the directory containing the Launchpad binary, run:

   ./launchpad connect grafana -k basic.config

2. Open the following URL in your browser to access Grafana: http://localhost:8080/d/rocketchat-metrics/rocket-chat-metrics.

3. Sign in with the default credentials:

   1. Username: admin

   2. Password: admin

   Grafana prompts you to change the password after your first login.

4. After signing in, the Rocket.Chat Metrics dashboard is displayed. The following screenshot shows an example:

   

5. To view the available dashboards, select Dashboards from the top menu.

   

6. Select Rocket.Chat Logs to view your workspace logs.

When you are finished, return to the terminal where the Grafana connection is running and press Ctrl + C to stop port forwarding. The Grafana dashboard is no longer accessible through the local URL.

Update your workspace

When updating your workspace using Launchpad, it updates the workspace custom resource in Kubernetes and monitors the rollout until all components are healthy.

To update a workspace to a new Rocket.Chat release, run:

./launchpad upgrade rocketchat -v <version>

Replace <version> with the Rocket.Chat version you want to update to, for example 8.6.0.

If your cluster has exactly one workspace, Launchpad finds it automatically. If you have more than one, specify the workspace name as an argument or with the -w flag:

./launchpad upgrade rocketchat <workspace-name> -v <version>

Here are some additional options you can use when updating your workspace version:

Update air-gapped workspace version

In air-gapped environments, Launchpad detects your private registry and verifies that all required container images are present before it applies any changes. If you need to push a new package bundle before upgrading, pass the bundle path and target registry following this command:

./launchpad upgrade rocketchat <workspace-name> \
  -v <version> \
  --package <package> \
  --registry <registry>

Scale your workspace

Use launchpad scale rocketchat to change the number of replicas for the Rocket.Chat core, DDP streamer, account, authorization, and presence services.

Built-in MongoDB and SeaweedFS are not scaled by this command. Their resources and storage remain based on the preset used during initial deployment.

Before applying a change, Launchpad checks that the new pods fit on your nodes. If they don't, the command stops and your workspace is not changed.

1. If you're scaling up to a larger preset, resize your cluster to the values in cluster sizing guide. On EKS, add nodes to your node group:

   eksctl scale nodegroup --cluster=dev-eks-launchpad --name=dev-eks-node-group --nodes=5 --nodes-max=5

   If the preset needs a larger instance type, create a new node group with that type instead.

2. Preview the change:

   ./launchpad scale rocketchat <workspace-name> --size medium --dry-run -k basic.config

   Launchpad shows the current and target replicas for each service.

3. Apply the change:

   ./launchpad scale rocketchat <workspace-name> --size medium -k basic.config

   Launchpad updates the workspace and waits until the new pods are running.

To scale down, run the same command with a smaller preset or lower counts. Users connected to removed pods reconnect automatically.

If the rollout fails, Launchpad prints a command that restores your previous replica counts. Run it, then check your cluster capacity.

Access Rocket.Chat Cloud

With Launchpad version 0.3.0, you can log in to the Rocket.Chat Cloud portal and take actions directly from the command line. The following commands are available:

• ./launchpad cloud login: Run this command to log into the Rocket.Chat Cloud portal (cloud.rocket.chat) from the CLI. You are prompted to authorize the login on a browser. This allows the Launchpad tool to perform actions on behalf of your Rocket.Chat Cloud account, which allows you to view and connect to new workspaces.

• ./launchpad cloud logout: Run this command to log out of the Cloud portal.

Troubleshooting

export KUBECONFIG=basic.config

$env:KUBECONFIG="basic.config"