Official Oracle MySQL Solution

MySQL Router Load Balancing

Deploy official MySQL Router for seamless InnoDB Cluster integration, intelligent connection routing, and enterprise-grade high availability with Oracle support.

Get Started Today View Comparative Matrix

MySQL Router implementation scope

MySQL Router provides application routing and InnoDB Cluster metadata discovery. JusDB configures read/write endpoints, supported connection-sharing options, TLS, and routing health checks. We test primary failure, metadata refresh, and application reconnect behavior against your workload's recovery objectives; automatic query splitting depends on the Router version and configuration.

Routing Ports: 6446 (RW) / 6447 (RO)·Recovery: Tested against your workload·Topology Sync: Dynamic Metadata Cache·Security: TLS 1.3 End-to-End·Response Targets: Per Agreement

R/W

Split Routing

Tested

Recovery Plan

6446/6447

Native Ports

MySQL Router Architecture with InnoDB Cluster

Comprehensive MySQL Router Services

From initial setup to advanced optimization, we provide complete MySQL Router solutions for InnoDB Cluster environments.

MySQL Router Implementation

Complete setup and configuration of MySQL Router for InnoDB Cluster integration

Router installation and configuration
InnoDB Cluster discovery setup
Connection routing configuration
SSL/TLS encryption setup

InnoDB Cluster Integration

Seamless integration with MySQL InnoDB Cluster for automatic failover

Cluster metadata management
Automatic topology discovery
Primary/secondary detection
Cluster state monitoring

Read/Write Splitting

Intelligent routing of read and write operations for optimal performance

Automatic read/write detection
Load balancing across secondaries
Connection pooling optimization
Transaction consistency handling

High Availability Setup

Automatic failover and recovery with minimal downtime and data loss

Automatic failover detection
Connection re-routing
Health monitoring and alerts
Graceful degradation handling

Performance Optimization

Fine-tune MySQL Router for maximum throughput and minimal latency

Connection pool tuning
Routing algorithm optimization
Memory and CPU optimization
Performance monitoring setup

Migration & Consulting

Expert guidance for migrating to MySQL Router and ongoing support

Legacy system migration
Architecture design review
Best practices training
Ongoing support and maintenance

Why Choose MySQL Router for InnoDB Cluster?

MySQL Router provides official Oracle support with seamless InnoDB Cluster integration and intelligent connection management.

Official Oracle Solution

Backed by Oracle with official support, regular updates, and seamless integration with MySQL ecosystem.

Intelligent Routing

Automatic read/write splitting with intelligent load balancing across secondary nodes for optimal performance.

Automatic Failover

Seamless failover with automatic detection of cluster topology changes and connection re-routing.

InnoDB Cluster Native

Purpose-built for InnoDB Cluster with native metadata management and cluster state awareness.

MySQL Router Routing Modes

Choose the optimal routing strategy for your application requirements and performance needs.

Read-Write Split

Automatic routing based on SQL statement type

Writes go to primary on port 6446, reads distributed across secondaries on port 6447 for optimal load distribution.

Best for: Mixed workloads

Read-Only

All connections routed to secondary nodes

Perfect for reporting, analytics, and read-heavy applications that don't require write access.

Best for: Analytics workloads

Primary-Only

All connections directed to the primary node

Ensures strong consistency for applications requiring immediate read-after-write consistency.

Best for: Strong consistency

Our MySQL Router Implementation Process

A systematic approach to deploying production-ready MySQL Router with comprehensive testing and optimization.

Cluster Assessment

Analyze your InnoDB Cluster setup, application requirements, and performance objectives.

Router Configuration

Install and configure MySQL Router with optimal settings for your cluster topology.

Integration Testing

Comprehensive testing of routing, failover scenarios, and application compatibility.

Production Deployment

Deploy to production with monitoring, alerting, and ongoing optimization support.

How JusDB MySQL Router compares to alternative routing proxies.

MySQL Router uses InnoDB Cluster metadata to select routing destinations. Read/write splitting depends on the configured endpoints and Router version; recovery depends on cluster behavior and application retries. We review connection sharing, socket limits, TLS, and failure tests against the workload requirements.

Swipe horizontally to compare MySQL Router engineering vectors→

Vector	JusDB	Cloud Native Default	Traditional Agency	In-House / DIY
Dynamic InnoDB Cluster Metadata Cache Synchronization	Real-time metadata polling and dynamic HTTP/REST cluster cache synchronization; automatically tracks primary promotions and replica quarantine states in sub-second cycles.	Static DNS endpoints or cloud managed proxies that poll every 15–30 seconds, leading to stale routing windows during membership changes.	Static HAProxy configuration requiring manual reload scripts on topology changes, causing socket drops and temporary routing black holes.	Uncoordinated config files manually synchronized across application hosts, leading to configuration drift and traffic routed to dead nodes.
Transparent Read/Write Splitting on Distinct Ports (6446/6447)	Native port separation (6446 RW classic, 6447 RO classic, 64460/64470 X Protocol) with round-robin secondary distribution and lag-aware read quarantine.	Single generic endpoint requiring application-level ORM read/write splitting configuration and manual replica target weighting.	Advises complex application-side database drivers with dual DataSource pooling, increasing connection churn and maintenance complexity.	Applications route all queries (reads and writes) to the primary node, causing CPU saturation while expensive read replicas sit idle.
Sub-Second Primary Failover Re-routing with Client Resiliency	Sub-second TCP connection termination and redirection to newly elected Group Replication primary, with client reconnect backoff handling and transaction isolation.	DNS TTL caching delays failover by 30–120 seconds; client libraries hang on stale TCP sockets until application socket timeout triggers.	Manual VIP failover or load balancer reconfiguration taking 5–15 minutes, accompanied by connection storms and API timeouts.	Application requests fail continuously with MySQL server has gone away errors until engineers manually intervene and update DNS.
Multi-Process Redundancy & TCP Connection Pooling	Co-located Router sidecar architecture or highly available Router tiers behind keepalived/VRRP, with optimized thread pool sizing and file descriptor tuning.	Single proxy instance per cluster or black-box managed proxy with rigid connection concurrency limits and hidden throttling.	Single centralized proxy node representing a catastrophic single point of failure (SPOF) for all application database traffic.	Direct unmanaged database connections hitting max_connections limits during traffic bursts, crashing the MySQL daemon.
TLS 1.3 End-to-End Encryption & Ephemeral Handshakes	End-to-end TLS 1.3 configuration with ephemeral Diffie-Hellman keys, automated certificate rotation, mTLS client verification, and zero handshake latency penalty.	Basic TLS terminated at cloud load balancer with unencrypted transit between proxy and backend DB instances inside VPC.	Self-signed legacy TLS 1.1/1.2 certificates with manual renewal procedures that expire unexpectedly and halt production connectivity.	Plaintext TCP connections over internal networks or disabled SSL verification flags (verify_server_cert=off) violating compliance mandates.
Initial-Response Targets & Senior DBRE Escalation	Named DBREs in a dedicated Slack war room, and proactive socket telemetry monitoring. Initial-response targets: 2 to 4 hours in covered business hours; 30 minutes for P1 incidents with 24x7 on-call. Targets, not resolution guarantees.	Cloud provider support tickets routed through automated tiers with 1–4 hour initial response time from non-database specialists.	Business-hours-only ticket queue with 4–8 hour SLAs; proxy configuration changes treated as billable change requests.	Application engineers forced to troubleshoot low-level proxy socket timeouts and network packet drops without deep DBRE expertise.

MySQL Router: Critical Failure Modes

Dynamic metadata polling, TCP connection handshakes, and replica lag introduce subtle failure boundaries under high client concurrency. Here are 3 mission-critical production failure modes our DBRE team mitigates:

P1 Critical

Stale Metadata Cache Routing Writes to Demoted Primary

During sudden primary failover, Router can temporarily serve cached cluster metadata if TTL or cluster contact intervals are loose. Applications continue sending writes to the demoted, read-only node, producing ER_OPTION_PREVENTS_STATEMENT error storms.

JusDB Engineering Mitigation:

JusDB tunes ttl parameter to sub-second thresholds, configures strict router client socket timeouts, and implements persistent dynamic metadata refresh loops with cluster VIP fallback.

P1 Critical

Connection Storm Saturating Max Client Descriptors

A flood of new application threads or backend reconnects after transient network drops exhausts max_connections or OS open file descriptors (ulimit -n), causing Router to drop incoming TCP SYNs and stall application ingress.

JusDB Engineering Mitigation:

We configure systemd resource limits, tune client_connect_timeout, size Router thread-stack buffers, and implement front-end rate limiting combined with backend persistent connection multiplexing.

P2 High

Read Replica Latency Causing Stale Read Inconsistencies

In asynchronous read replica pools, high replication lag on port 6447 secondaries results in read-after-write anomalies, where users execute updates on port 6446 but immediately read stale snapshots on port 6447.

JusDB Engineering Mitigation:

JusDB configures dynamic read quarantine based on replication lag thresholds, routes critical read-your-writes transactions to port 6446, and tunes Group Replication flow control parameters.

Telemetry Runbooks · Non-Blocking MySQL Router Diagnostics

Our MySQL DBREs execute non-blocking diagnostics to inspect Router daemon version, port listener bindings, and active HTTP REST routing endpoints:

CLI: Router Binary Version & Listening Ports

Shell · Network Verification

Verifies MySQL Router binary build version and inspects kernel TCP socket listeners for ports 6446 (read-write) and 6447 (read-only).

# 1. Verify MySQL Router binary version & build details
mysqlrouter --version

# 2. Inspect active TCP listener ports for classic & X-protocol
netstat -tulpn | grep 6446

HTTP REST API: Active Routes & Destination Health

REST API · Route Telemetry

Queries MySQL Router REST API to inspect active route endpoints, destination backend nodes, and active client connection counts.

# 1. Query Router REST API for active routing endpoints
curl -s http://localhost:8081/api/v1/routes | jq .

# 2. Inspect active client connections and destination weights
curl -s http://localhost:8081/api/v1/routes/bootstrap_rw/destinations | jq .

MySQL Router & InnoDB Cluster FAQs

Common questions about MySQL Router implementation and InnoDB Cluster integration.

What makes MySQL Router different from other MySQL load balancers?

MySQL Router is the official Oracle solution specifically designed for InnoDB Cluster. It provides native cluster metadata integration, automatic topology discovery, and seamless failover without requiring external configuration management.

How does MySQL Router handle read/write splitting?

MySQL Router supports separate read/write and read-only endpoints. MySQL Router 8.4 also supports automatic read/write splitting when access_mode=auto, connection_sharing=1, the classic protocol, and a PRIMARY_AND_SECONDARY metadata-cache destination are configured. We validate transaction behavior, read-after-write requirements, and application compatibility before enabling it.

Can MySQL Router work with existing applications?

Yes, MySQL Router acts as a transparent proxy, requiring minimal application changes. Applications connect to Router as if connecting directly to MySQL, making migration straightforward with immediate high availability benefits.

How fast is failover with MySQL Router?

Recovery time depends on cluster failure detection, primary election, metadata refresh, routing configuration, and application reconnect behavior. We test the complete failure path against your recovery objective rather than promise a universal failover time.

What monitoring capabilities does MySQL Router provide?

MySQL Router provides comprehensive logging, performance metrics, connection statistics, and health status information. It integrates with MySQL Enterprise Monitor and supports custom monitoring solutions through REST API endpoints.

Is MySQL Router suitable for high-traffic applications?

Yes, MySQL Router is designed for enterprise workloads and can handle thousands of concurrent connections with minimal latency overhead. It supports connection pooling and can be deployed in multiple instances for horizontal scaling.

Related MySQL Services

Explore complementary MySQL high availability and clustering services that pair with MySQL Router.

MySQL InnoDB Cluster

Built-in high availability with Group Replication and MySQL Shell

Deploy and manage MySQL InnoDB Cluster — the native topology MySQL Router routes connections to.

Explore MySQL InnoDB Cluster services

MySQL High Availability

Resilient MySQL architectures with automatic failover

Design fault-tolerant MySQL deployments with minimal downtime and seamless failover.

Explore MySQL high availability

MySQL Orchestrator

Topology discovery and automated failover for replication fleets

Topology management and automated recovery for asynchronous MySQL replication topologies.

Explore MySQL Orchestrator services

Ready to Implement MySQL Router?

Let our experts deploy official MySQL Router solutions for your InnoDB Cluster environment.

Start Your Project Request a Consultation