How to Configure ScyllaDB Proxy

Published
Length
3 min read

Luis Felipe Dias

All posts

You set up a scylladb proxy by routing client traffic through an intermediate network host. Next, you map that traffic to native database roles and TLS rules. Engineering teams often start with a basic load balancer or direct node routing. They create shared database accounts so apps and engineers can read and write data. Over time, these shared roles gain broad permissions across many keyspaces. A single mistake or leaked password then puts the whole cluster at risk. How a native scylladb pro

Your coding agents run with your access.

Claude Code reaches prod with a developer's credentials, and approval prompts get clicked through. hoop.dev checks every query, masks sensitive data, and holds risky changes for a human.

You set up a scylladb proxy by routing client traffic through an intermediate network host. Next, you map that traffic to native database roles and TLS rules. Engineering teams often start with a basic load balancer or direct node routing. They create shared database accounts so apps and engineers can read and write data. Over time, these shared roles gain broad permissions across many keyspaces. A single mistake or leaked password then puts the whole cluster at risk.

How a native scylladb proxy and role setup works

ScyllaDB uses a peer-to-peer ring design where all nodes are equal. Clients talk to any node in the cluster using the CQL binary protocol. That chosen node acts as the coordinator for the current request. It sends the read or write query to the nodes holding the actual data.

Some teams place a network load balancer in front of these database nodes. This setup gives apps a single address to reach the database cluster. Still, client drivers often bypass this middle hop after the first login. The driver learns the cluster topology and talks straight to each node instead. Because of this, admins must lock down security on every single database node.

Inside the database, admins turn on password checks and role-based rules. They create specific roles for different teams, apps, and service accounts. Then they grant read, write, or alter rights on target keyspaces and tables. You also set up TLS encryption to protect data moving across the network. Each node checks the client certificate or password before allowing a session.

What native roles solve and what they miss

Native database roles and network rules solve the basic login problem. They stop unknown users from reaching your data nodes on the network. They also separate production app traffic from basic read-only analytics users. When a developer leaves the team, you can drop their personal database role. This setup keeps unauthorized people out of your keyspaces.

Still, native access rules only check who logs in at the start. They do not inspect what individual queries do while the session runs. If a support engineer has read rights on a customer table, they see everything. Queries will return raw emails, phone numbers, card numbers, and national IDs. Native table grants cannot mask those sensitive fields on the fly. They must allow the whole column or block the query completely.

Database Proxy (ProxySQL, PgBouncer) + End-to-End Encryption: architecture patterns and best practices

Continue reading? Get the full guide.

Free. No spam. Unsubscribe anytime.

Broad write roles also create big risks during live fixing sessions. An engineer with modify rights can drop a partition or overwrite good rows. Native logs often miss the link between a shared role and a specific person. Teams also struggle to manage static passwords stored on developer laptops. Rotating these passwords across dozens of scripts and apps takes hours of manual work.

Why network routing alone falls short for data teams

Standard network routers only see raw TCP packets moving between hosts. They do not understand the CQL wire protocol or the queries inside it. A simple load balancer cannot tell a safe read from a full table scan. It just forwards bytes from the client to the database node.

When teams scale, managing firewall rules for every developer IP becomes a chore. Remote workers change networks often, which breaks strict IP allowlists right away. Teams then open up wider network ranges just to get daily work done. This habit weakens your security posture and exposes database ports to more threats. To fix this, you need controls that understand both user identity and query content.

Add runtime guardrails and masking beside your databases and APIs

Login rules, role grants, and CI checks govern who connects to your stack. Yet any allowed script, engineer, or AI agent still runs commands with the full scope of its role. For workloads on Postgres, MySQL, SQL Server, MongoDB, ClickHouse, Kubernetes, SSH, or internal HTTP services, hoop.dev sits inline in the protocol beside the service as an open-source proxy, a sidecar for your database or API. On every production query an agent or human runs, hoop.dev decides whether to pass, mask, block, or escalate it for review.

Before a query runs, hoop.dev guardrails check the statement against your rules, an AI Analyzer you configure can score intent or hold risky calls for human approval, and hoop.dev data masking rewrites sensitive fields in responses in memory. Explore the MIT-licensed project on GitHub.

Frequently asked questions

Why do CQL client drivers bypass standard network load balancers?

Token-aware drivers read the cluster map during the first network handshake. After that, they send queries straight to the nodes that hold the data.

Can native ScyllaDB roles hide sensitive columns from support engineers?

No, native permissions apply to whole tables or keyspaces at a time. They cannot rewrite query results to mask emails or card numbers in flight.

How do teams audit who ran a query when using shared database roles?

Native audit logs only record the shared database role name and client IP. You need an identity-linked control point to tie each query to a real user.

The query runs. The emails stay hidden.