Deploying a ScyllaDB App

Introduction

ScyllaDB is a high-performance, Cassandra-compatible NoSQL database. Deploying ScyllaDB with a Dockerfile on Klutch.sh provides reproducible builds, managed secrets, and persistent storage—all configured from klutch.sh/app. This guide focuses on a single-node deployment using one TCP app for the CQL port.

---

Prerequisites

• A Klutch.sh account (sign up)
• A GitHub repository containing your ScyllaDB Dockerfile (GitHub is the only supported git source)
• A CQL client (e.g., cqlsh) to test connectivity

For onboarding, see the Quick Start.

---

Architecture and ports

• Klutch.sh allows one port per app. Expose ScyllaDB’s CQL TCP port only:
  • CQL: TCP traffic on internal port 9042; clients connect via example-app.klutch.sh:8000 (Klutch TCP external) mapped to internal 9042.
• This guide uses a single-node setup; clustering requires additional ports and is not covered here due to single-port routing.

---

Repository layout

scylladb/
├── Dockerfile              # Must be at repo root for auto-detection
└── README.md

Keep secrets out of Git; store them in Klutch.sh environment variables.

---

Installation (local) and starter commands

Build and run locally:

docker build -t scylla-local .
docker run -p 9042:9042 \
  -e SCYLLA_CLUSTER_NAME=klutch-scylla \
  -e SCYLLA_SEEDS=127.0.0.1 \
  scylla-local --smp 1 --memory 1G --overprovisioned 1

---

Dockerfile for ScyllaDB (production-ready)

Place this at the repo root; Klutch.sh auto-detects Dockerfiles.

FROM scylladb/scylla:latest

EXPOSE 9042

# Run with conservative defaults for single-node in constrained environments
CMD ["--smp", "1", "--memory", "1G", "--overprovisioned", "1"]

Notes:

• Pin to a specific tag (e.g., scylladb/scylla:5.4) for stability.
• Adjust --memory and --smp to match your plan’s resources.

---

Environment variables (Klutch.sh)

Common settings:

• SCYLLA_CLUSTER_NAME=klutch-scylla
• SCYLLA_SEEDS=127.0.0.1 (single-node)
• Optional: SCYLLA_ENDPOINT_SNITCH=GossipingPropertyFileSnitch

If deploying without the Dockerfile and relying on Nixpacks:

• NIXPACKS_START_CMD=./scylla --smp 1 --memory 1G --overprovisioned 1

---

Attach persistent volumes

Add storage in Klutch.sh (path and size only):

• /var/lib/scylla — data files and commit logs.

Ensure the path is writable inside the container.

---

Deploy ScyllaDB on Klutch.sh (Dockerfile workflow)

1. Push your repository—with the Dockerfile at the root—to GitHub.
2. Open klutch.sh/app, create a project, and add an app.
3. Select TCP traffic and set the internal port to 9042.
4. Add environment variables (cluster name, seeds) as needed.
5. Attach a volume at /var/lib/scylla sized for your data.
6. Deploy. Connect with a CQL client at example-app.klutch.sh:8000.

---

Sample CQL usage

Create a keyspace and table:

CREATE KEYSPACE IF NOT EXISTS demo WITH replication = {'class': 'SimpleStrategy', 'replication_factor': 1};
USE demo;
CREATE TABLE IF NOT EXISTS users (id uuid PRIMARY KEY, name text);
INSERT INTO users (id, name) VALUES (uuid(), 'Alice');
SELECT * FROM users;

Run these statements via cqlsh example-app.klutch.sh 8000.

---

Health checks and production tips

• Use nodetool status (from a maintenance pod/host) to verify node health.
• Keep single-node constraints in mind; clustering requires additional ports not exposed here.
• Pin image versions and test upgrades in staging before production.
• Monitor volume usage on /var/lib/scylla; resize before hitting capacity.

---

ScyllaDB on Klutch.sh delivers a reproducible single-node setup with managed secrets and persistent storage—without extra YAML or CI steps. Configure the TCP port, attach storage, and connect your clients over CQL to start serving data.