MySQL Fabric

MySQL Fabric
Public domain · source
Name	MySQL Fabric
Developer	Oracle Corporation
Released	2013
Latest release	1.5.5
Programming language	Python (programming language)
Operating system	Linux, Windows, macOS
Platform	x86, ARM architecture
License	GNU General Public License

MySQL Fabric MySQL Fabric is a centralized management framework created to provide high-availability and sharding services for the MySQL ecosystem. It was developed by Oracle Corporation engineers and integrated with tools commonly used alongside Percona distributions and MariaDB forks in enterprise deployments. Fabric aimed to coordinate topology, failover, and partitioning across clusters managed by administrators using interfaces compatible with Ansible, Puppet, and Chef automation.

Overview

MySQL Fabric provided two primary modes: high-availability group management and sharding orchestration, designed for deployments similar to patterns used by Facebook, Twitter, LinkedIn, and Netflix for scalable storage. It targeted operational models implemented in datacenters like those run by Amazon Web Services, Google Cloud Platform, Microsoft Azure, and hosted offerings from Rackspace. Fabric’s control plane was inspired by concepts used in Apache Zookeeper, etcd, and orchestration systems such as Kubernetes and Docker Swarm, while integrating with SQL ecosystems including PostgreSQL in heterogeneous architectures.

Architecture

The Fabric architecture centered on a daemon process written in Python (programming language) that coordinated metadata stored in a persistent repository; this design paralleled metadata services such as Consul and Apache Cassandra metadata layers. Components included the Fabric daemon, coordinator nodes, and managed MySQL server instances configured as replication groups resembling topologies used by Galera Cluster and Semi-synchronous replication setups. Communication used protocols and clients similar to those in MySQL Replication and management APIs reminiscent of Prometheus exporters and Nagios check plugins. Fabric exposed control interfaces enabling integration with administration tools like MySQL Workbench and command-line utilities found in distributions from Oracle Corporation and third-party vendors like Percona.

Features

Fabric implemented automated group membership, leader election, and online failover workflows comparable to mechanisms in Paxos and Raft-based systems used by projects such as HashiCorp Consul and etcd. It supported sharding through a directory of shard maps and routing hints akin to approaches used by Vitess and Citus (extension), and provided APIs for resharding operations similar to rebalancing in Hadoop Distributed File System clusters. Operational features included automated promotion of replicas, maintenance modes, and scripted hooks that could interoperate with orchestration and CI/CD pipelines from Jenkins, Travis CI, and GitLab CI/CD. Fabric’s design emphasized integration with enterprise identity systems such as LDAP and Active Directory for administrative control.

Deployment and Configuration

Typical deployments used Fabric alongside configuration management systems like Ansible, Puppet, and Chef to provision servers on platforms such as Amazon EC2, Google Compute Engine, and private infrastructure from Dell EMC or Hewlett Packard Enterprise. Configuration involved defining server groups, replication topologies, and shard definitions through Fabric’s CLI and configuration files, paralleling practices used in OpenStack and OpenShift provisioning. Administrators often paired Fabric with monitoring stacks using Grafana, Prometheus, and log aggregation via ELK Stack components including Elasticsearch and Logstash.

Operational Management and Monitoring

Fabric exposed operational controls for failover, switchover, and topology changes that integrated with alerting systems like PagerDuty and Opsgenie. Monitoring best practices involved collecting metrics from MySQL instances and the Fabric daemon using exporters compatible with Prometheus and dashboards modelled after deployments described in case studies by Facebook Engineering and Twitter Engineering. Backup workflows invoked utilities similar to mysqldump and logical/physical backup tools popularized by Percona XtraBackup, while maintenance automation used scripts and orchestration patterns illustrated in Kubernetes operator designs.

Limitations and Deprecation

Despite its capabilities, Fabric had limitations in scalability and active development compared to other ecosystem projects; these constraints mirrored challenges encountered by orchestration efforts such as earlier versions of OpenStack and some Hadoop subprojects. Oracle shifted priorities, and Fabric’s maintenance cadence slowed as attention moved to alternate solutions including InnoDB Cluster, Group Replication, Vitess, and third-party offerings from Percona and MariaDB Corporation. This transition resulted in deprecation signals from vendor roadmaps and community forks referencing long-term maintenance efforts exemplified by migrations from deprecated services like Google Reader and discontinued OpenSolaris components.

Community and Development History

Development of Fabric began under the stewardship of teams at Oracle Corporation after the acquisition of Sun Microsystems, and its early releases were accompanied by presentations at conferences such as Oracle OpenWorld and community events like Percona Live. Community contributions and forks occurred in repositories maintained by individuals and organizations active in the MySQL ecosystem, including contributors from Percona, MariaDB Corporation, and independent operators who documented migration paths to alternatives like Vitess and InnoDB Cluster. Discussions about Fabric’s roadmap appeared in forums and mailing lists frequented by participants from Stack Overflow, GitHub, and developer communities around Linux Foundation projects.

Category:Database management systems