About

• 1: Features
• 2: History
• 3: News & Events
• 4: Roadmap
• 5: Join the Community
• 6: Privacy Policy
• 7: License
• 8: Sponsor Us
• 9: User Cases
• 10: Subscription
• 11: FAQ
• 12: Release Note
• 13: Comparison
  • 13.1: Cost Reference
  • 13.2: Open-Source Impact

1 - Features

“PostgreSQL In Great STYle”: Postgres, Infras, Graphics, Service, Toolbox, it’s all Yours.

—— Battery-included, local-first PostgreSQL distribution, open-source RDS alternative

Value Propositions

• Extensibility: Powerful extensions out of the box: deep integration of PostGIS, TimescaleDB, Citus, PGVector, 576 plugins, and 12 PG kernels.
• Reliability: Quickly create high-availability, self-healing PostgreSQL clusters with built-in point-in-time recovery, access control, a self-signed CA, and TLS.
• Observability: Based on Prometheus & Grafana modern observability stack, providing stunning monitoring best practices. Modular design, can be used independently: Gallery & Demo.
• Availability: Deliver stable, reliable, auto-routed, transaction-pooled, read-write separated high-performance database services, with flexible access modes via HAProxy, Pgbouncer, and VIP.
• Maintainability: Easy to use, Infrastructure as Code, Management SOPs, auto-tuning, local software repository, Vagrant sandbox and Terraform templates, zero-downtime migration solutions.
• Composability: Modular architecture design, reusable Infra, various optional modules: Redis, Silo object storage, ETCD, DuckDB, Docker, Supabase.

Overview

Pigsty is a better local open-source RDS for PostgreSQL alternative:

• Battery-Included RDS: From kernel to RDS distribution, providing production-grade PG database services for versions 14-18 on EL/Debian/Ubuntu.
• Rich Extensions: Providing unparalleled 576 extensions with out-of-the-box distributed, time-series, geospatial, graph, vector, multi-modal database capabilities.
• Flexible Modular Architecture: Compose Redis, Etcd, and Silo object-storage modules with PostgreSQL modes such as Mongo; monitor existing RDS, hosts, and databases independently.
• Stunning Observability: Based on modern observability stack Prometheus/Grafana, providing stunning, unparalleled database observability capabilities.
• Battle-Tested Reliability: Self-healing high-availability architecture: automatic failover on hardware failure, seamless traffic switching. With auto-configured PITR as safety net for accidental data deletion!
• Easy to Use and Maintain: Declarative API, GitOps ready, foolproof operation, Database/Infra-as-Code and management SOPs encapsulating management complexity!
• Solid Security Practices: HBA, ACL, TLS, backup, logging, and host-firewall foundations, with explicit default boundaries and production hardening requirements.
• Broad Application Scenarios: Low-code data application development, or use preset Docker Compose templates to spin up massive software using PostgreSQL with one click!
• Open-Source Free Software: Own better database services at less than 1/10 the cost of cloud databases! Truly “own” your data and achieve autonomy!

PostgreSQL integrates ecosystem tools and best practices:

• Out-of-the-box PostgreSQL distribution, deeply integrating 576 packaged extensions for geospatial, time-series, distributed, graph, vector, search, and AI!
• Runs on bare operating systems without container support, supporting mainstream operating systems: EL 8/9/10, Ubuntu 22.04/24.04/26.04, and Debian 12/13.
• Based on patroni, haproxy, and etcd, creating a self-healing high-availability architecture: automatic failover on hardware failure, seamless traffic switching.
• Combines pgBackRest with optional Silo object storage to provide out-of-the-box point-in-time recovery (PITR), protecting against software defects and accidental data deletion.
• Based on Ansible providing declarative APIs to abstract complexity, greatly simplifying daily operations management in a Database-as-Code manner.
• Pigsty has broad applications, can be used as complete application runtime, develop demo data/visualization applications, and massive software using PG can be spun up with Docker templates.
• Provides Vagrant-based local development and testing sandbox environment, and Terraform-based cloud auto-deployment solutions, keeping development, testing, and production environments consistent.
• Run PostgreSQL in Mongo-compatible mode with DocumentDB and the FerretDB Docker APP

Battery-Included RDS

Get production-grade PostgreSQL database services locally immediately!

PostgreSQL is a near-perfect database kernel, but it needs more tools and systems to become a good enough database service (RDS). Pigsty helps PostgreSQL make this leap. Pigsty solves various challenges you’ll encounter when using PostgreSQL: kernel extension installation, connection pooling, load balancing, service access, high availability / automatic failover, log collection, metrics monitoring, alerting, backup recovery, PITR, access control, parameter tuning, security encryption, certificate issuance, NTP, DNS, parameter tuning, configuration management, CMDB, management playbooks… You no longer need to worry about these details!

Pigsty supports PostgreSQL 14 ~ 18 mainline kernels and other compatible forks, running on EL / Debian / Ubuntu and compatible OS distributions, available on x86_64 and ARM64 chip architectures, without container support required. Besides database kernels and many out-of-the-box extension plugins, Pigsty also provides complete infrastructure and runtime required for database services, as well as local sandbox / production environment / cloud IaaS auto-deployment solutions.

Pigsty can bootstrap an entire environment from bare metal with one click, reaching the last mile of software delivery. Ordinary developers and operations engineers can quickly get started and manage databases part-time, building enterprise-grade RDS services without database experts!

Rich Extensions

Hyper-converged multi-modal, use PostgreSQL for everything, one PG to replace all databases!

PostgreSQL’s soul lies in its rich extension ecosystem, and Pigsty uniquely deeply integrates 576 extensions from the PostgreSQL ecosystem, providing you with an out-of-the-box hyper-converged multi-modal database!

Extensions can create synergistic effects, producing 1+1 far greater than 2 results. You can use PostGIS for geospatial data, TimescaleDB for time-series/event stream data analysis, and Citus to upgrade it in-place to a distributed geospatial-temporal database; You can use PGVector to store and search AI embeddings, ParadeDB for ElasticSearch-level full-text search, and simultaneously use precise SQL, full-text search, and fuzzy vector for hybrid search. You can also achieve dedicated OLAP database/data lakehouse analytical performance through pg_duckdb, pg_mooncake and other analytical extensions.

Using PostgreSQL as a single component to replace MySQL, Kafka, ElasticSearch, MongoDB, and big data analytics stacks has become a best practice — a single database choice can significantly reduce system complexity, greatly improve development efficiency and agility, achieving remarkable software/hardware and development/operations cost reduction and efficiency improvement.

Flexible Modular Architecture

Flexible composition, free extension, multi-database support, monitor existing RDS/hosts/databases

Components in Pigsty are abstracted as independently deployable modules, which can be freely combined to address varying requirements. The INFRA module comes with a complete modern monitoring stack, while the NODE module tunes nodes to desired state and brings them under management. Installing the PGSQL module on multiple nodes automatically forms a high-availability database cluster based on primary-replica replication, while the ETCD module provides consensus and metadata storage for database high availability.

Beyond these four core modules, Pigsty also provides a series of optional feature modules: The MINIO module can deploy Silo to provide local object storage and serve as a centralized database backup repository. The REDIS module can provide auxiliary services for databases in standalone primary-replica, sentinel, or native cluster modes. The DOCKER module can be used to spin up stateless application software.

Additionally, Pigsty provides PG-compatible / derivative kernel support. You can use Babelfish for MS SQL Server compatibility, IvorySQL for Oracle compatibility, OpenHaloDB for MySQL compatibility, and OrioleDB for ultimate OLTP performance.

Furthermore, you can use PostgreSQL Mongo mode for MongoDB compatibility, Supabase for Firebase compatibility, and PolarDB to meet domestic compliance requirements. Message queues are covered by the KAFKA module, which deploys Kafka 4.x dynamic KRaft clusters. More professional/pilot modules will be continuously introduced to Pigsty, such as GPSQL, DUCKDB, TIGERBEETLE, KUBERNETES, CONSUL, GREENPLUM, CLOUDBERRY, MYSQL, …

Stunning Observability

Using modern open-source observability stack, providing unparalleled monitoring best practices!

Pigsty provides best practices for monitoring based on the open-source Grafana / Prometheus modern observability stack: Grafana for visualization, VictoriaMetrics for metrics collection, VictoriaLogs for log collection and querying, Alertmanager for alert notifications. Blackbox Exporter for checking service availability. The entire system is also designed for one-click deployment as the out-of-the-box INFRA module.

Pigsty automatically monitors every managed component: host nodes, HAProxy load balancers, PostgreSQL databases, PgBouncer connection pools, Etcd metadata stores, Redis-compatible caches, Silo object storage, and the monitoring infrastructure itself. Grafana dashboards and preset alert rules provide immediate operational visibility. The same stack can also monitor applications, existing database instances, and cloud RDS services.

Whether for failure analysis or slow query optimization, capacity assessment or resource planning, Pigsty provides comprehensive data support, truly achieving data-driven operations. In Pigsty, over three thousand types of monitoring metrics are used to describe all aspects of the entire system, and are further processed, aggregated, analyzed, refined, and presented in intuitive visualization modes. From global overview dashboards to CRUD details of individual objects (tables, indexes, functions) in a database instance, everything is visible at a glance. You can drill down, roll up, or jump horizontally freely, browsing current system status and historical trends, and predicting future evolution.

Additionally, Pigsty’s monitoring system module can be used independently — to monitor existing host nodes and database instances, or cloud RDS services. With just one connection string and one command, you can get the ultimate PostgreSQL observability experience.

Visit the Screenshot Gallery and Online Demo for more details.

Battle-Tested Reliability

Out-of-the-box high availability and point-in-time recovery capabilities ensure your database is rock-solid!

For table/database drops caused by software defects or human error, Pigsty provides out-of-the-box PITR point-in-time recovery capability, enabled by default without additional configuration. As long as storage space allows, base backups and WAL archiving based on pgBackRest let you quickly return to any point within the recovery window. You can use local directories/disks, Silo deployed by the MINIO module, or external S3-compatible object-storage services to retain longer recovery windows, according to your budget.

Pigsty provides a high-availability self-healing architecture based on Patroni, etcd, and HAProxy. When the node, network, quorum, and synchronous-replica assumptions hold, it can fail over the primary automatically. Actual RTO and RPO depend on replication mode, failure type, timeout settings, and client reconnection behavior.

Pigsty includes built-in HAProxy load balancers for automatic traffic switching, providing DNS/VIP/LVS and other access methods for clients. Failover and active switchover are almost imperceptible to the business side except for brief interruptions, and applications don’t need to modify connection strings or restart. The minimal maintenance window requirements bring great flexibility and convenience: you can perform rolling maintenance and upgrades on the entire cluster without application coordination. The feature that hardware failures can wait until the next day to handle lets developers, operations, and DBAs sleep well. Many large organizations and core institutions have been using Pigsty in production for extended periods. The largest deployment has 25K CPU cores and 200+ PostgreSQL ultra-large instances; in this deployment case, dozens of hardware failures and various incidents occurred over six to seven years, DBAs changed several times, but still maintained availability higher than 99.999%.

Easy to Use and Maintain

Infra as Code, Database as Code, declarative APIs encapsulate database management complexity.

Pigsty provides services through declarative interfaces, elevating system controllability to a new level: users tell Pigsty “what kind of database cluster I want” through configuration inventories, without worrying about how to do it. In effect, this is similar to CRDs and Operators in K8S, but Pigsty can be used for databases and infrastructure on any node: whether containers, virtual machines, or physical machines.

Whether creating/destroying clusters, adding/removing replicas, or creating new databases/users/services/extensions/whitelist rules, you only need to modify the configuration inventory and run the idempotent playbooks provided by Pigsty, and Pigsty adjusts the system to your desired state. Users don’t need to worry about configuration details — Pigsty automatically tunes based on machine hardware configuration. You only need to care about basics like cluster name, how many instances on which machines, what configuration template to use: transaction/analytics/critical/tiny — developers can also self-serve. But if you’re willing to dive into the rabbit hole, Pigsty also provides rich and fine-grained control parameters to meet the demanding customization needs of the most meticulous DBAs.

Beyond that, Pigsty’s own installation and deployment is also one-click foolproof, with all dependencies pre-packaged, requiring no internet access during installation. The machine resources needed for installation can also be automatically obtained through Vagrant or Terraform templates, allowing you to spin up a complete Pigsty deployment from scratch on a local laptop or cloud VM in about ten minutes. The local sandbox environment can run on a 1-core 2GB micro VM, providing the same functional simulation as production environments, usable for development, testing, demos, and learning.

Solid Security Practices

Pigsty provides the security foundations required for database deployment: layered HBA, built-in roles and default privileges, SCRAM-SHA-256, page checksums, a local CA, component certificates, backup, PITR, centralized logs, and firewall configuration.

The defaults target development, testing, and demonstrations on a trusted intranet. Production deployments must replace public credentials, review network boundaries, enforce TLS where required, configure server-certificate verification, and establish backup recovery, privilege review, and incident-response processes.

Security and Compliance documents each mechanism’s default state and boundary. Security Considerations provides production hardening guidance, and Compliance maps relevant controls to MLPS and SOC 2. Whether a deployment meets a specific requirement depends on scope, organizational process, continuous evidence, and the auditor’s conclusion.

Broad Application Scenarios

Use preset Docker templates to spin up massive software using PostgreSQL with one click!

In various data-intensive applications, the database is often the trickiest part. For example, the core difference between GitLab Enterprise and Community Edition is the underlying PostgreSQL database monitoring and high availability. If you already have a good enough local PG RDS, you can refuse to pay for software’s homemade database components.

Pigsty provides the Docker module and many out-of-the-box Compose templates. You can use Pigsty-managed high-availability PostgreSQL (as well as Redis and Silo) as backend storage, spinning up these software in stateless mode with one click: GitLab, Gitea, Wiki.js, NocoDB, Odoo, Jira, Confluence, Harbor, Mastodon, Discourse, KeyCloak, Mattermost, etc. If your application needs a reliable PostgreSQL database, Pigsty is perhaps the simplest way to get one.

Pigsty also provides application development toolsets closely related to PostgreSQL: PGAdmin4, PGWeb, ByteBase, PostgREST, Kong, as well as EdgeDB, FerretDB, Supabase — these “upper-layer databases” using PostgreSQL as storage. More wonderfully, you can build interactive data applications quickly in a low-code manner based on the Grafana and Postgres built into Pigsty, and even use Pigsty’s built-in ECharts panels to create more expressive interactive visualization works.

Pigsty provides a powerful runtime for your AI applications. Your agents can leverage PostgreSQL and the powerful capabilities of the observability world in this environment to quickly build data-driven intelligent agents.

Open-Source Free Software

Pigsty is free software open-sourced under Apache-2.0, watered by the passion of PostgreSQL-loving community members

Pigsty is completely open-source and free software, allowing you to run enterprise-grade PostgreSQL database services at nearly pure hardware cost without database experts. For comparison, database vendors’ “enterprise database services” and public cloud vendors’ RDS charge premiums several to over ten times the underlying hardware resources as “service fees.”

Many users choose the cloud precisely because they can’t handle databases themselves; many users use RDS because there’s no other choice. We will break cloud vendors’ monopoly, providing users with a cloud-neutral, better open-source RDS alternative: Pigsty follows PostgreSQL upstream closely, with no vendor lock-in, no annoying “licensing fees,” no node count limits, and no data collection. All your core assets — data — can be “autonomously controlled,” in your own hands.

Pigsty itself aims to replace tedious manual database operations with database autopilot software, but even the best software can’t solve all problems. There will always be some rare, low-frequency edge cases requiring expert intervention. This is why we also provide professional subscription services to provide safety nets for enterprise users who need them. Subscription consulting fees of tens of thousands are less than one-thirtieth of a top DBA’s annual salary, completely eliminating your concerns and putting costs where they really matter. For community users, we also contribute with love, providing free support and daily Q&A.

tooltip: { trigger: axis, formatter: $fn:ttfmt }
legend: { top: 4, itemGap: 16, data: [Oracle, Open-Source PG, Cloud RDS, Pigsty over IaaS, Pigsty over IDC ] }
grid: { left: 96, right: 36, bottom: 70, top: 50 }
xAxis:
  type: category
  name: CPU Cores
  nameLocation: middle
  nameGap: 36
  boundaryGap: false
  data: [2, 4, 8, 12, 16, 24, 32, 52, 64, 104, 128, 196, 256, 384, 512]
yAxis:
  type: log
  logBase: 10
  min: 10
  name: Monthly Cost (CNY)
  axisLabel: { formatter: $fn:yfmt }
  splitLine: { show: true, lineStyle: { type: dashed, opacity: 0.5 } }
series:
  - { name: Oracle, type: line, symbolSize: 7, lineStyle: { width: 3 }, itemStyle: { color: "#d62728" }, data: [45000, 65000, 105000, 145000, 185000, 265000, 345000, 545000, 665000, 1065000, 1305000, 1985000, 2585000, 3865000, 5145000] }
  - { name: Cloud RDS, type: line, symbolSize: 6, lineStyle: { width: 2 }, itemStyle: { color: "#ff7f0e" }, data: [800, 1600, 3200, 4800, 6400, 9600, 12800, 20800, 25600, 41600, 51200, 78400, 102400, 153600, 204800] }
  - { name: Pigsty over IaaS, type: line, symbolSize: 6, lineStyle: { width: 2 }, itemStyle: { color: "#2ca02c" }, data: [360, 720, 1440, 2160, 2880, 4320, 5760, 9360, 11520, 18720, 23040, 35280, 46080, 69120, 92160] }
  - { name: Pigsty over IDC, type: line, symbolSize: 6, lineStyle: { width: 2 }, itemStyle: { color: "#9467bd" }, data: [38, 76, 152, 228, 304, 456, 608, 988, 1216, 1976, 2432, 3724, 4864, 7296, 9728] }

2 - History

Historical Origins

The Pigsty project began in 2018-2019, originating from Tantan. Tantan is an internet dating app — China’s Tinder, now acquired by Momo. Tantan was a Nordic-style startup with a Swedish engineering founding team.

Tantan had excellent technical taste, using PostgreSQL and Go as its core technology stack. The entire Tantan system architecture was modeled after Instagram, designed entirely around the PostgreSQL database. Up to several million daily active users, millions of TPS, and hundreds of TB of data, the data component used only PostgreSQL. Almost all business logic was implemented using PG stored procedures — even including 100ms recommendation algorithms! It was arguably the most complex PostgreSQL-at-scale use case in China at the time.

This atypical development model of deeply using PostgreSQL features placed extremely high demands on the capabilities of engineers and DBAs. And Pigsty is the open-source project we forged in this real-world large-scale, high-standard database cluster scenario — embodying our experience and best practices as top PostgreSQL experts.

Development Process

In the beginning, Pigsty did not have the vision, goals, and scope it has today. It started as a PostgreSQL monitoring system for our own use. We surveyed all available solutions — open-source, commercial, cloud-based, datadog, pgwatch, etc. — and none could meet our observability needs. So I decided to build one myself based on Grafana and Prometheus. This became Pigsty’s predecessor and prototype. Pigsty as a monitoring system was quite impressive, helping us solve countless management problems.

Subsequently, developers wanted such a monitoring system on their local development machines, so we used Ansible to write provisioning playbooks, transforming this system from a one-time construction task into reusable, replicable software. New versions allowed users to use Vagrant and Terraform, using Infrastructure as Code to quickly spin up local DevBox development machines or production environment servers, automatically completing PostgreSQL and monitoring system deployment.

Next, we redesigned the production environment PostgreSQL architecture, introducing Patroni and pgBackRest to solve database high availability and point-in-time recovery issues. We developed a zero-downtime migration solution based on logical replication, rolling upgrading two hundred production database clusters to the latest major version through blue-green deployment. And we incorporated these capabilities into Pigsty.

Pigsty is software we built for ourselves. The biggest benefit of “eating our own dog food” is that we are both developers and users — as client users, we know exactly what we need, do not cut corners, and never worry about automating ourselves out of jobs.

We solved problem after problem, depositing the solutions into Pigsty. Pigsty’s positioning also gradually evolved from a monitoring system into an out-of-the-box PostgreSQL database distribution. We then decided to open-source Pigsty and began a series of technical sharing and publicity, and external users from various industries began using Pigsty and providing feedback.

Full-Time Entrepreneurship

In 2022, the Pigsty project received seed funding from Miracle Plus, initiated by Dr. Qi Lu, allowing me to work on this full-time.

As an open-source project, Pigsty has developed quite well. In these years of full-time work, Pigsty’s GitHub stars grew from a few hundred to 5,213 as of 2026-07-11; it made the HN front page, and growth began snowballing. In November 2025, Pigsty won the Magneto Award at the PostgreSQL Ecosystem Conference. In 2026, Pigsty’s subproject PGEXT.CLOUD was selected for a PGCon.Dev 2026 talk. Pigsty became the first Chinese open-source project to appear on the stage of this core PostgreSQL ecosystem conference.

Previously, Pigsty could only run on CentOS 7, but now it covers all mainstream Linux distributions (EL, Debian, Ubuntu) across 16 operating system platforms. Supported PG major versions cover 14-18, and we maintain and integrate 576 extension plugins in the PG ecosystem. Among these, I personally maintain over half (360+) of the extension plugins, providing out-of-the-box RPM/DEB packages. Including Pigsty itself, “based on open source, giving back to open source,” this is our way of contributing to the PG ecosystem.

Pigsty’s positioning has also continuously evolved from a PostgreSQL database distribution to an open-source cloud database. It truly benchmarks against cloud vendors’ entire cloud database brands.

Rebel Against Public Clouds

Public cloud vendors like AWS, Azure, GCP, and Aliyun have provided many conveniences for startups, but they are closed-source and force users to rent infrastructure at exorbitant fees.

We believe that excellent database services, like excellent database kernels, should be accessible to every user, rather than requiring expensive rental from cyber lords.

Cloud computing’s agility and elasticity value proposition is strong, but it should be free, open-source, inclusive, and local-first — We believe the cloud computing universe needs a solution representing open-source values that returns infrastructure control to users without sacrificing the benefits of the cloud.

Therefore, we are also leading a movement and battle to exit the cloud, as rebels against public clouds, to reshape the industry’s values.

Our Vision

I hope that in the future world, everyone will have the de facto right to freely use excellent services, rather than being confined to a few cyber lord public cloud giants’ territories as cyber tenants or even cyber serfs.

This is exactly what Pigsty aims to do — a better, free and open-source RDS alternative. Allowing users to spin up database services better than cloud RDS anywhere (including cloud servers) with one click.

Pigsty is a complete complement to PostgreSQL, and a spicy mockery of cloud databases. It literally means “pigsty,” but it’s also an acronym for Postgres In Great STYle, meaning “PostgreSQL in its full glory.”

Pigsty itself is completely open-source and free software, so you can build a PostgreSQL service that scores 90 without database experts. We sustain operations by providing premium consulting services to take you from 90 to 100, with warranty, Q&A, and a safety net.

A well-built system may run for years without needing a “safety net,” but database problems, once they occur, are never small. Often, expert experience can turn decay into magic, and we provide such premium consulting — we believe this is a more just, reasonable, and sustainable model.

About the Team

I am Feng Ruohang, the author of Pigsty. Almost all of Pigsty’s code is developed by me alone.

Individual heroism still exists in the software field. Only unique individuals can create unique works — I hope Pigsty becomes such a work.

If you’re interested in me, here’s my personal homepage: https://vonng.com/

“Modb Interview with Feng Ruohang” (Chinese)

“Post-90s, Quit to Start Business, Says Will Crush Cloud Databases” (Chinese)

3 - News & Events

Recent News

• 2026-08-15: Pigsty v4.5.0 officially released: Kafka, MySQL, Valkey, Silo, and 575 extensions

  • Release Blog: Pigsty v4.5 Release Article
  • Release Notes: v4.5.0
  • Highlights: Kafka KRaft and MySQL 8.4 pilot modules, Valkey in REDIS, Silo as the MINIO backend, 575 total extensions, cluster-identity-aware orchestration, and offline packages for all seven recommended OS versions.

• 2026-07-10: Pigsty v4.4.0 officially released: PG19 beta support and 531 extensions

  • Release Notes: v4.4.0
  • Highlights: PostgreSQL 18.4 / 19 beta support, 531 total extensions, kernel and package updates, and Pig CLI improvements.

• 2026-05-01: Pigsty v4.3.0 officially released: 510 extensions, Ubuntu 26 support

  • Release Notes: v4.3.0
  • Highlights: Infra / PGSQL / kernel package batch updates, Ubuntu 26.04 offline package coverage, and 510 total extensions.

• 2026-03-06: Pigsty v4.2.1 released! Drop PG13, 464 extensions

  • Release Notes: v4.2.1

• 2026-02-28: Pigsty v4.2 is officially released! Seven kernel updates shipped together

  • Release Blog: Pigsty v4.2 Release Article
  • Release Notes: v4.2.0

• 2026-02-12: Pigsty v4.1 is officially released! First distribution batch with PostgreSQL 18.2 support

  • Release Blog: Pigsty v4.1 Release Article
  • Release Notes: v4.1.0
  • Pigsty minor-release support is now available: 18.2…

• 2026-02-04: Extension for Everyone selected as a PGCon.Dev 2026 talk!

  • The first Chinese vendor on the PGConf.Dev keynote stage

• 2026-02-03: Pigsty v4.0 Released! Entering the Agent era!

  • PostgreSQL official news: “Pigsty v4.0 Released: Ready for the Agent Era”
  • Victoria observability revolution, security hardening, new JUICE/VIBE modules, container support, license switched to Apache-2.0
  • Release Notes: v4.0.0
  • Release Blog: Pigsty v4.0: Entering the AI Era

• 2026-01-30: PIG v1.0 Released! Launched together with the PGEXT.CLOUD extension catalog

  • PostgreSQL official news: “PIG v1.0 Released with PGEXT.CLOUD: 444 PG extensions on 14 Linux”
  • PostgreSQL extension package manager pig v1.0 is GA, with PGEXT.CLOUD open extension infrastructure and 444 extensions

• 2025-12-02: Pigsty v3.7.0 Released! PG18 becomes default, 437 extensions, EL10/Debian13 support

  • Release Notes: v3.7.0

• 2025-11-29: Pigsty won the PostgreSQL Magneto Award!

  • The 8th Conference of PostgreSQL Ecosystem (Hangzhou, China)
  • Topics: “A World-Grade Postgres Meta Distribution”, AI database considerations, PostgreSQL delivery best practices

• 2025-08-15: Pigsty v3.6.1 Released! Routine PG minor update, PGDG China regional mirrors

  • Release Notes: v3.6.1

• 2025-08-04: Pigsty v3.6.0 Released! PostgreSQL meta-distribution

  • PostgreSQL official news: “Pigsty 3.6, the meta-distribution for PostgreSQL”
  • pgactive multi-primary replication, MinIO/ETCD improvements, simplified installation, configuration cleanup
  • Release Notes: v3.6.0

• 2025-06-16: Pigsty v3.5.0 Released! PG18 beta support, 421 extensions, monitoring upgrade, code refactor

  • Release Notes: v3.5.0

• 2025-04-21: Pigsty v3.4 Released! MySQL compatibility

  • PostgreSQL official news: “Pigsty v3.4 Released, PG RDS with MySQL Compatibility”
  • OpenHalo/OrioleDB support, better backups, automatic Certbot certificates, AGE extension
  • Release Notes: v3.4.1 / v3.4.0

• 2025-03-07: Pigsty v3.3.0 Released! 404 extensions

  • PostgreSQL official news: “Pigsty v3.3 Release: with 404 PostgreSQL Extensions”
  • Odoo/Dify/Supabase app templates, DocumentDB support
  • Release Notes: v3.3.0

• 2025-01: Pigsty v3.2.x release series (v3.2.0 ~ v3.2.2)

  • 390 extensions, Omnigres, Mooncake, Citus13/PG17, extension management CLI, Grafana improvements, ARM64 extension coverage
  • Release Notes: v3.2.2 / v3.2.1 / v3.2.0

• PostgreSQL Package Manager pig Released!

  • ANNOUNCE pig: The Postgres Extension Wizard
  • PIG: A New Package Manager for PostgreSQL Extensions

• 2024-11: Pigsty v3.1.0 Released! PG17 default, self-hosted Supabase, ARM/Ubuntu24 support

  • PostgreSQL official news: “Pigsty v3.1 Release: PG17, Duck Extensions, Self-hosting Supabase, ARM & Ubuntu24”
  • Blog Post: “Pigsty v3.1: Self-hosted Supabase, PG17 Default, MinIO Improvements, ARM & Ubuntu24 Support”
  • Release Notes: v3.1.0

• 2024-08 ~ 2024-10: Pigsty v3.0.x release series (v3.0.0 ~ v3.0.4)

  • PostgreSQL official news: “Pigsty v3: 336 extensions and MSSQL/Oracle flavor PG kernels!”
  • 333 extensions, replaceable kernels, MSSQL/Oracle/PolarDB compatibility, PG17 extensions
  • Release Notes: v3.0.0 ~ v3.0.4
  • Feature Introduction: Pigsty v3.0.0

• 2024-08: Pigsty supplementary repository provides 254 additional ready-to-use binary RPM/DEB extensions!

  • PostgreSQL official news: “Pigsty Supplementary APT/YUM Repository with 254 additional PostgreSQL Extensions!”

• PGCon.Dev 2024 Trip Report!

• 2024-05: Pigsty v2.7 Released!

  • PostgreSQL official news: “Pigsty v2.7 Released, free RDS PG with 255 extensions available”
  • Pigsty Blog: Pigsty v2.7: The Great Integration

• 2024-02: Pigsty v2.6 Released!

  • PostgreSQL official news: “Pigsty, Battery-included PostgreSQL Distro & Free RDS Alternative, v2.6 released!”
  • Pigsty Blog: Pigsty v2.6: PG Challenges OLAP

Conferences & Talks

4 - Roadmap

Release Strategy

Pigsty uses semantic versioning: <major>.<minor>.<patch>. Alpha/Beta/RC versions will have suffixes like -a1, -b1, -c1 appended to the version number.

Major version updates signify incompatible foundational changes and major new features; minor version updates typically indicate regular feature updates and small API changes; patch version updates mean bug fixes and package version updates.

Pigsty plans to release one major version update per year. Minor version updates usually follow PostgreSQL’s minor version update rhythm, catching up within a month at the latest after a new PostgreSQL version is released. Pigsty typically plans 4-6 minor versions per year. For complete release history, please refer to Release Notes.

Features Under Consideration

• Agent Native CLI - PIG
• DBA Agent - basic integration
• Grafana dashboard improvements
• Boar management console

Here are our Active Issues and Roadmap.

Extensions and Packages

For the extension support roadmap, you can find it here: /ext/e/roadmap

Under Consideration

• PDU: https://github.com/wublabdubdub/PDU-PostgreSQLDataUnloader
• walminer https://gitee.com/movead/XLogMiner
• is_jsonb_valid https://github.com/furstenheim/is_jsonb_valid
• pg_kafka https://github.com/xstevens/pg_kafka
• pg_jieba https://github.com/jaiminpan/pg_jieba
• pg_paxos https://github.com/microsoft/pg_paxos
• OneSparse https://github.com/OneSparse/OneSparse
• PipelineDB https://github.com/pipelinedb/pipelinedb
• SQL Firewall https://github.com/uptimejp/sql_firewall
• zcurve https://github.com/bmuratshin/zcurve
• PG dot net https://github.com/Brick-Abode/pldotnet/releases
• pg_scws: https://github.com/jaiminpan/pg_scws
• themsis: https://github.com/cossacklabs/pg_themis
• pgspeck https://github.com/johto/pgspeck
• lsm3 https://github.com/postgrespro/lsm3
• monq https://github.com/postgrespro/monq
• pg_badplan https://github.com/trustly/pg_badplan
• pg_recall https://github.com/mreithub/pg_recall
• pgfsm https://github.com/michelp/pgfsm
• pg_trgm pro https://github.com/postgrespro/pg_trgm_pro
• pgsql-fio: https://github.com/csimsek/pgsql-fio

Not Considering for Now

• pg_tier: not ready due to incomplete dep parquet_s3_fdw
• parquet_s3_fdw: not ready due to compiler version
• pg_top: not ready due to cmake error
• timestamp9: not ready due to compiler error
• pg_tier obsolete
• pg_timeseries, we already have timescaledb
• pg_quack, we already have a pg_lakehouse
• pg_telemetry, we already have better observability
• pgx_ulid, https://github.com/pksunkara/pgx_ulid, already covered by pg_idkit (MIT, but RUST)
• embedding: obsolete
• FEAT zson https://github.com/postgrespro/zson MIT C (too old)
• GIS pghydro https://github.com/pghydro/pghydro C GPL-2.0 6.6 (no makefile)
• https://github.com/Zeleo/pg_natural_sort_order (too old)
• https://github.com/postgrespro/pg_query_state
• https://github.com/no0p/pgsampler
• pg_lz4 https://github.com/zilder/pg_lz4
• pg_amqp https://github.com/omniti-labs/pg_amqp
• tinyint https://github.com/umitanuki/tinyint-postgresql
• pg_blkchain https://github.com/blkchain/pg_blkchain
• hashtypes https://github.com/pandrewhk/hashtypes
• foreign_table_exposer https://github.com/komamitsu/foreign_table_exposer
• ldap_fdw https://github.com/guedes/ldap_fdw
• pg_backtrace https://github.com/postgrespro/pg_backtrace
• connection_limits https://github.com/tvondra/connection_limits
• fixeddecimal https://github.com/2ndQuadrant/fixeddecimal

5 - Join the Community

GitHub

Our GitHub repository is: https://github.com/pgsty/pigsty. Please give us a ⭐️ star!

We welcome anyone to submit new Issues or create Pull Requests, propose feature suggestions, and contribute to Pigsty.

Please note that for issues related to Pigsty documentation, please submit Issues in the github.com/pgsty/pigsty.cc repository.

Press ⌘+ with K on macOS, or Ctrl+ with K, to search the documentation, extension catalog, and blog directly.

Maintainers

Pigsty is built by its maintainers and community.

WeChat Groups

Chinese users are mainly active in WeChat groups. Currently, there are seven active groups. Groups 1-4 are full; for other groups, you need to add the assistant’s WeChat to be invited.

To join the WeChat community, search for “Pigsty小助手” (WeChat ID: pigsty-cc), note or send “加群” (join group), and the assistant will invite you to the group.

International Community

Telegram: https://t.me/joinchat/gV9zfZraNPM3YjFh

Discord: https://discord.gg/j5pG8qfKxU

You can also contact me via email: [email protected]

Community Help

When you encounter problems using Pigsty, you can seek help from the community. The more information you provide, the more likely you are to get help from the community.

Please refer to the Community Help Guide and provide as much information as possible so that community members can help you solve the problem. Here is a reference template for asking for help:

What happened? (Required)

Pigsty version and OS version (Required)

$ grep version pigsty.yml

$ cat /etc/os-release

$ uname -a

Some cloud providers have customized standard OS distributions. You can tell us which cloud provider’s OS image you are using. If you have customized and modified the environment after installing the OS, or if there are specific security rules and firewall configurations in your LAN, please also inform us when asking questions.

Pigsty configuration file

Please don’t forget to redact any sensitive information: passwords, internal keys, sensitive configurations, etc.

cat ~/pigsty/pigsty.yml

What did you expect to happen?

Please describe what should happen under normal circumstances, and how the actual situation differs from expectations.

How to reproduce this issue?

Please tell us in as much detail as possible how to reproduce this issue.

Monitoring screenshots

If you are using the monitoring system provided by Pigsty, you can provide relevant screenshots.

Error logs

Please provide logs related to the error as much as possible. Please do not paste content like “Failed to start xxx service” that has no informational value.

You can query logs from Grafana / VictoriaLogs, or get logs from the following locations:

• Syslog: /var/log/messages (rhel) or /var/log/syslog (debian)
• Postgres: /pg/log/postgres/*
• Patroni: /pg/log/patroni/*
• Pgbouncer: /pg/log/pgbouncer/*
• Pgbackrest: /pg/log/pgbackrest/*

journalctl -u patroni
journalctl -u <service name>

Have you searched Issues/Website/FAQ?

In the FAQ, we provide answers to many common questions. Please check before asking.

You can also search for related issues from GitHub Issues and Discussions:

• Pigsty FAQ
• Github Issues
• Pigsty Discussions

Is there any other information we need to know?

The more information and context you provide, the more likely we can help you solve the problem.

6 - Privacy Policy

Pigsty Software

When you install Pigsty software, if you use offline package installation in a network-isolated environment, we will not receive any data about you.

If you choose online installation, when downloading related packages, our servers or cloud provider servers will automatically log the visiting machine’s IP address and/or hostname in the logs, along with the package names you downloaded.

We will not share this information with other organizations unless required by law. (Honestly, we’d have to be really bored to look at this stuff.)

Pigsty’s primary domain is: pigsty.io. For mainland China, please use the registered mirror site pigsty.cc.

Pigsty Website

When you visit our website, our servers will automatically log your IP address and/or hostname in Nginx logs.

We will only store information such as your email address, name, and location when you decide to send us such information by completing a survey or registering as a user on one of our websites.

We collect this information to help us improve website content, customize web page layouts, and contact people for technical and support purposes. We will not share your email address with other organizations unless required by law.

This website uses Google Analytics, a web analytics service provided by Google, Inc. (“Google”). Google Analytics uses “cookies,” which are text files placed on your computer to help the website analyze how users use the site.

The information generated by the cookie about your use of the website (including your IP address) will be transmitted to and stored by Google on servers in the United States. Google will use this information to evaluate your use of the website, compile reports on website activity for website operators, and provide other services related to website activity and internet usage. Google may also transfer this information to third parties if required by law or where such third parties process the information on Google’s behalf. Google will not associate your IP address with any other data held by Google. You may refuse the use of cookies by selecting the appropriate settings on your browser, however, please note that if you do this, you may not be able to use the full functionality of this website. By using this website, you consent to the processing of data about you by Google in the manner and for the purposes set out above.

If you have any questions or comments about this policy, or request deletion of personal data, you can contact us by sending an email to [email protected]

7 - License

License Summary

Pigsty core uses Apache-2.0; documentation uses CC BY 4.0.

Official License: https://github.com/pgsty/pigsty/blob/main/LICENSE

Pigsty Core

The Pigsty core is licensed under Apache License 2.0.

Apache-2.0 is a permissive open-source license. You may freely use, modify, and distribute the software for commercial purposes without opening your own source code or adopting the same license.

Pigsty Documentation

Pigsty documentation sites (pigsty.cc, pigsty.io, pgsty.com) use Creative Commons Attribution 4.0 International (CC BY 4.0).

CC BY 4.0 permits free sharing and adaptation with appropriate credit, a license link, and indication of changes.

SBOM Inventory

Open-source software used or related to the Pigsty project.

For 576 PostgreSQL extension plugin licenses, refer to PostgreSQL Extension License List.

Necessity Levels:

• Required: Essential core capabilities, no option to disable
• Recommended: Enabled by default, can be disabled via configuration
• Optional: Not enabled by default, can be enabled via configuration

Apache-2.0 License Text

                                 Apache License
                           Version 2.0, January 2004
                        http://www.apache.org/licenses/

   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION

   1. Definitions.

      "License" shall mean the terms and conditions for use, reproduction,
      and distribution as defined by Sections 1 through 9 of this document.

      "Licensor" shall mean the copyright owner or entity authorized by
      the copyright owner that is granting the License.

      "Legal Entity" shall mean the union of the acting entity and all
      other entities that control, are controlled by, or are under common
      control with that entity. For the purposes of this definition,
      "control" means (i) the power, direct or indirect, to cause the
      direction or management of such entity, whether by contract or
      otherwise, or (ii) ownership of fifty percent (50%) or more of the
      outstanding shares, or (iii) beneficial ownership of such entity.

      "You" (or "Your") shall mean an individual or Legal Entity
      exercising permissions granted by this License.

      "Source" form shall mean the preferred form for making modifications,
      including but not limited to software source code, documentation
      source, and configuration files.

      "Object" form shall mean any form resulting from mechanical
      transformation or translation of a Source form, including but
      not limited to compiled object code, generated documentation,
      and conversions to other media types.

      "Work" shall mean the work of authorship, whether in Source or
      Object form, made available under the License, as indicated by a
      copyright notice that is included in or attached to the work
      (an example is provided in the Appendix below).

      "Derivative Works" shall mean any work, whether in Source or Object
      form, that is based on (or derived from) the Work and for which the
      editorial revisions, annotations, elaborations, or other modifications
      represent, as a whole, an original work of authorship. For the purposes
      of this License, Derivative Works shall not include works that remain
      separable from, or merely link (or bind by name) to the interfaces of,
      the Work and Derivative Works thereof.

      "Contribution" shall mean any work of authorship, including
      the original version of the Work and any modifications or additions
      to that Work or Derivative Works thereof, that is intentionally
      submitted to Licensor for inclusion in the Work by the copyright owner
      or by an individual or Legal Entity authorized to submit on behalf of
      the copyright owner. For the purposes of this definition, "submitted"
      means any form of electronic, verbal, or written communication sent
      to the Licensor or its representatives, including but not limited to
      communication on electronic mailing lists, source code control systems,
      and issue tracking systems that are managed by, or on behalf of, the
      Licensor for the purpose of discussing and improving the Work, but
      excluding communication that is conspicuously marked or otherwise
      designated in writing by the copyright owner as "Not a Contribution."

      "Contributor" shall mean Licensor and any individual or Legal Entity
      on behalf of whom a Contribution has been received by Licensor and
      subsequently incorporated within the Work.

   2. Grant of Copyright License. Subject to the terms and conditions of
      this License, each Contributor hereby grants to You a perpetual,
      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
      copyright license to reproduce, prepare Derivative Works of,
      publicly display, publicly perform, sublicense, and distribute the
      Work and such Derivative Works in Source or Object form.

   3. Grant of Patent License. Subject to the terms and conditions of
      this License, each Contributor hereby grants to You a perpetual,
      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
      (except as stated in this section) patent license to make, have made,
      use, offer to sell, sell, import, and otherwise transfer the Work,
      where such license applies only to those patent claims licensable
      by such Contributor that are necessarily infringed by their
      Contribution(s) alone or by combination of their Contribution(s)
      with the Work to which such Contribution(s) was submitted. If You
      institute patent litigation against any entity (including a
      cross-claim or counterclaim in a lawsuit) alleging that the Work
      or a Contribution incorporated within the Work constitutes direct
      or contributory patent infringement, then any patent licenses
      granted to You under this License for that Work shall terminate
      as of the date such litigation is filed.

   4. Redistribution. You may reproduce and distribute copies of the
      Work or Derivative Works thereof in any medium, with or without
      modifications, and in Source or Object form, provided that You
      meet the following conditions:

      (a) You must give any other recipients of the Work or
          Derivative Works a copy of this License; and

      (b) You must cause any modified files to carry prominent notices
          stating that You changed the files; and

      (c) You must retain, in the Source form of any Derivative Works
          that You distribute, all copyright, patent, trademark, and
          attribution notices from the Source form of the Work,
          excluding those notices that do not pertain to any part of
          the Derivative Works; and

      (d) If the Work includes a "NOTICE" text file as part of its
          distribution, then any Derivative Works that You distribute must
          include a readable copy of the attribution notices contained
          within such NOTICE file, excluding those notices that do not
          pertain to any part of the Derivative Works, in at least one
          of the following places: within a NOTICE text file distributed
          as part of the Derivative Works; within the Source form or
          documentation, if provided along with the Derivative Works; or,
          within a display generated by the Derivative Works, if and
          wherever such third-party notices normally appear. The contents
          of the NOTICE file are for informational purposes only and
          do not modify the License. You may add Your own attribution
          notices within Derivative Works that You distribute, alongside
          or as an addendum to the NOTICE text from the Work, provided
          that such additional attribution notices cannot be construed
          as modifying the License.

      You may add Your own copyright statement to Your modifications and
      may provide additional or different license terms and conditions
      for use, reproduction, or distribution of Your modifications, or
      for any such Derivative Works as a whole, provided Your use,
      reproduction, and distribution of the Work otherwise complies with
      the conditions stated in this License.

   5. Submission of Contributions. Unless You explicitly state otherwise,
      any Contribution intentionally submitted for inclusion in the Work
      by You to the Licensor shall be under the terms and conditions of
      this License, without any additional terms or conditions.
      Notwithstanding the above, nothing herein shall supersede or modify
      the terms of any separate license agreement you may have executed
      with Licensor regarding such Contributions.

   6. Trademarks. This License does not grant permission to use the trade
      names, trademarks, service marks, or product names of the Licensor,
      except as required for reasonable and customary use in describing the
      origin of the Work and reproducing the content of the NOTICE file.

   7. Disclaimer of Warranty. Unless required by applicable law or
      agreed to in writing, Licensor provides the Work (and each
      Contributor provides its Contributions) on an "AS IS" BASIS,
      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
      implied, including, without limitation, any warranties or conditions
      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
      PARTICULAR PURPOSE. You are solely responsible for determining the
      appropriateness of using or redistributing the Work and assume any
      risks associated with Your exercise of permissions under this License.

   8. Limitation of Liability. In no event and under no legal theory,
      whether in tort (including negligence), contract, or otherwise,
      unless required by applicable law (such as deliberate and grossly
      negligent acts) or agreed to in writing, shall any Contributor be
      liable to You for damages, including any direct, indirect, special,
      incidental, or consequential damages of any character arising as a
      result of this License or out of the use or inability to use the
      Work (including but not limited to damages for loss of goodwill,
      work stoppage, computer failure or malfunction, or any and all
      other commercial damages or losses), even if such Contributor
      has been advised of the possibility of such damages.

   9. Accepting Warranty or Additional Liability. While redistributing
      the Work or Derivative Works thereof, You may choose to offer,
      and charge a fee for, acceptance of support, warranty, indemnity,
      or other liability obligations and/or rights consistent with this
      License. However, in accepting such obligations, You may act only
      on Your own behalf and on Your sole responsibility, not on behalf
      of any other Contributor, and only if You agree to indemnify,
      defend, and hold each Contributor harmless for any liability
      incurred by, or claims asserted against, such Contributor by reason
      of your accepting any such warranty or additional liability.

   END OF TERMS AND CONDITIONS

   APPENDIX: How to apply the Apache License to your work.

      To apply the Apache License to your work, attach the following
      boilerplate notice, with the fields enclosed by brackets "[]"
      replaced with your own identifying information. (Don't include
      the brackets!)  The text should be enclosed in the appropriate
      comment syntax for the file format. We also recommend that a
      file or class name and description of purpose be included on the
      same "printed page" as the copyright notice for easier
      identification within third-party archives.

   Copyright (C) 2018-2026  Ruohang Feng, @Vonng ([email protected])

   Licensed under the Apache License, Version 2.0 (the "License");
   you may not use this file except in compliance with the License.
   You may obtain a copy of the License at

       http://www.apache.org/licenses/LICENSE-2.0

   Unless required by applicable law or agreed to in writing, software
   distributed under the License is distributed on an "AS IS" BASIS,
   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
   See the License for the specific language governing permissions and
   limitations under the License.

8 - Sponsor Us

Sponsor Us

Pigsty is a free and open-source software, passionately developed by PostgreSQL community members, aiming to integrate the power of the PostgreSQL ecosystem and promote the widespread adoption of PostgreSQL. If our work has helped you, please consider sponsoring or supporting our project:

• Sponsor us directly with financial support - express your sincere support in the most direct and powerful way!
• Consider purchasing our Technical Support Services. We can provide professional PostgreSQL high-availability cluster deployment and maintenance services, making your budget worthwhile!
• Share your Pigsty use cases and experiences through articles, talks, and videos.
• Allow us to mention your organization in “Users of Pigsty.”
• Recommend/refer our project and services to friends, colleagues, and clients in need.
• Follow our WeChat Official Account and share relevant technical articles to groups and your social media.

Angel Investors

Pigsty is a project invested by Miracle Plus (formerly YC China) S22. We thank Miracle Plus and Dr. Qi Lu for their support of this project!

Sponsors

Special thanks to Vercel for sponsoring pigsty and hosting the Pigsty website.

Special thanks to JetBrains for sponsoring Pigsty with JetBrains Open Source License

9 - User Cases

According to Google Analytics PV and download statistics, Pigsty currently has approximately 100,000 users, with half from mainland China and half from other regions globally. They span across multiple industries including internet, cloud computing, finance, autonomous driving, manufacturing, tech innovation, ISV, and defense. If you are using Pigsty and are willing to share your case and Logo with us, please contact us - we offer one free consultation session as a token of appreciation.

Internet

Tantan: 200+ physical machines for PostgreSQL and Redis services

Bilibili: Supporting PostgreSQL innovative business

Cloud Vendors

Bitdeer: Providing PG DBaaS

Oracle OCI: Using Pigsty to deliver PostgreSQL clusters

Finance

AirWallex: Monitoring 200+ GCP PostgreSQL databases

Media & Entertainment

Media Storm: Self-hosted PG RDS / Victoria Metrics

Autonomous Driving

Momenta: Autonomous driving, managing self-hosted PostgreSQL clusters

Manufacturing

Huafon Group: Using Pigsty to deliver PostgreSQL clusters as chemical industry time-series data warehouse

Tech Innovation

Beijing Lingwu Technology: Migrating PostgreSQL from cloud to self-hosted

Motphys: Self-hosted PostgreSQL supporting GitLab

Sailong Biotech: Self-hosted Supabase

Hangzhou Lingma Technology: Self-hosted PostgreSQL

ISV

Inner Mongolia Haode Tianmu Technology Co., Ltd.

Shanghai Yuanfang

DSG

10 - Subscription

Pigsty aims to unite the power of the PostgreSQL ecosystem and help users make the most of the world’s most popular database, PostgreSQL, with self-driving database management software.

While Pigsty itself has already resolved many issues in PostgreSQL usage, achieving truly enterprise-grade service quality requires expert support and comprehensive coverage from the original provider. We deeply understand the importance of professional commercial support for enterprise customers. Therefore, Pigsty Enterprise Edition provides a series of value-added services on top of the open-source version, helping users better utilize PostgreSQL and Pigsty for customers to choose according to their needs.

If you have any of the following needs, please consider Pigsty subscription service:

• Running databases in critical scenarios requiring strict SLA guarantees and comprehensive coverage.
• Need comprehensive support for complex issues related to Pigsty and PostgreSQL.
• Seeking guidance on PostgreSQL/Pigsty production environment best practices.
• Want experts to help interpret monitoring dashboards, analyze and identify performance bottlenecks and fault root causes, and provide recommendations.
• Need to plan database architectures that meet security/disaster recovery/compliance requirements based on existing resources and business needs.
• Need to migrate from other databases to PostgreSQL, or migrate and transform legacy instances.
• Building an observability system, data dashboards, and visualization applications based on the Victoria/Grafana technology stack.
• Migrating off cloud and seeking open-source alternatives to RDS for PostgreSQL - cloud-neutral, vendor lock-in-free solutions.
• Want professional support for Redis/ETCD/Silo, as well as extensions like TimescaleDB/Citus.
• Want to perform secondary development and OEM branding with explicit commercial authorization.
• Want to sell Pigsty as SaaS/PaaS/DBaaS, or provide technical services/consulting/cloud services based on this distribution.

Subscription Plans

In addition to the Open Source Edition, Pigsty offers two different subscription service tiers: Professional Edition and Enterprise Edition, which you can choose based on your actual situation and needs.

Note on /price: The /price page is a simplified global pricing landing page (USD pricing, includes the Standard tier and node-cap presets). This page is the detailed subscription reference (CNY pricing, delivery scope, and OS/PG compatibility matrix). For technical compatibility boundaries, this page and Supported Linux prevail.

Pigsty Open Source Edition (OSS)

Pigsty Open Source Edition uses the Apache-2.0 license, provides complete core functionality, requires no fees, but does not guarantee any warranty service. If you find defects in Pigsty, we welcome you to submit an Issue on Github.

Pigsty Open Source supports seven currently validated baselines: EL 9.8 / 10.2, Debian 12.15 / 13.6, and Ubuntu 22.04.5 / 24.04.4 / 26.04.0, across both x86_64 and aarch64. v4.5.0 publishes a dual-architecture offline bundle for each of those seven baselines, fourteen artifacts in total, all freely downloadable; see the offline installation guide.

Using the Pigsty open source version allows junior development/operations engineers to have 70%+ of the capabilities of professional DBAs. Even without database experts, they can easily set up a highly available, high-performance, easy-to-maintain, secure and reliable PostgreSQL database cluster.

= Primary support, = Optional support

Pigsty Professional Edition (PRO)

Pigsty Professional Edition subscription provides complete functional modules and warranty for Pigsty itself. For defects in PostgreSQL itself and extension plugins, we will make our best efforts to provide feedback and fixes through the PostgreSQL global developer community.

Pigsty Professional Edition is built on the open source version, fully compatible with all open source features, and provides additional modules plus broader database/OS compatibility options: we provide build options for all minor versions of eight mainstream Linux releases (EL8/9/10, Debian 12/13, Ubuntu 22/24/26).

Pigsty Professional Edition includes support for PostgreSQL 14 - 18, and tracks upstream PostgreSQL minor updates continuously (for active majors, typically day-zero or near-day availability), ensuring smooth rolling upgrades to newer majors and minors.

Pigsty Professional Edition subscription allows you to use China mainland mirror site software repositories, accessible without VPN/proxy; we will also customize offline software installation packages for your exact operating system major/minor version, ensuring normal installation and delivery in air-gapped environments, achieving autonomous and controllable deployment.

Pigsty Professional Edition subscription provides standard expert consulting services, including complex issue analysis, DBA Q&A support, backup compliance advice, etc. We commit to responding to your issues within business hours (5x8), and provide 1 person-day support per year, with optional person-day add-on options.

Pigsty Professional Edition uses a commercial license, providing additional modules, technical support, and warranty services.

Pigsty Professional Edition starting price is ¥150,000 / year, equivalent to the annual fee for 9 vCPU AWS high-availability RDS PostgreSQL, or a junior operations engineer with a monthly salary of 10,000 yuan.

Pigsty Enterprise Edition

Pigsty Enterprise Edition subscription includes all service content provided by the Pigsty Professional Edition subscription, plus the following value-added service items:

Pigsty Enterprise Edition subscription provides the broadest range of database/operating system version support, including extended support for EOL operating systems (EL7, D11), domestic operating systems, cloud vendor operating systems, and legacy PostgreSQL major versions (PG12+ on request), as well as full support for Arm64 architecture chips.

Pigsty Enterprise Edition subscription provides domestic innovation and localization solutions, allowing you to use PolarDB v2.0 (this kernel license needs to be purchased separately) kernel to replace the native PostgreSQL kernel and meet local compliance requirements.

Pigsty Enterprise Edition subscription provides higher-standard enterprise-level consulting services, committing to 7x24 with (< 1h) response time SLA, and can provide more types of consulting support: version upgrades, performance bottleneck identification, annual architecture review, extension plugin integration, etc.

Pigsty Enterprise Edition subscription includes 2 person-days of support per year, with optional person-day add-on options, for resolving more complex and time-consuming issues.

Pigsty Enterprise Edition allows you to use Pigsty for DBaaS purposes, building cloud database services for external sales.

Pigsty Enterprise Edition starting price is ¥400,000 / year, equivalent to the annual fee for 24 vCPU AWS high-availability RDS, or an operations expert with a monthly salary of 30,000 yuan.

Pigsty Subscription Notes

Feature Differences

Pigsty Professional/Enterprise Edition includes the following additional features compared to the open source version:

• Command Line Management Tool: Unlock the full functionality of the Pigsty command line tool (pig)
• System Customization Capability: Provide pre-built offline installation packages for exact mainstream Linux operating system distribution major/minor versions
• Offline Installation Capability: Complete Pigsty installation in environments without Internet access (air-gapped environments)
• Multi-version PG Kernel: Allow users to freely specify and install PostgreSQL major versions within the lifecycle (14 - 18)
• Kernel Replacement Capability: Allow users to use other PostgreSQL-compatible kernels to replace the native PG kernel, and the ability to install these kernels offline
  • Babelfish: Provides Microsoft SQL Server wire protocol-level compatibility
  • IvorySQL: Based on PG, provides Oracle syntax/type/stored procedure compatibility
  • PolarDB PG: Provides support for open-source PolarDB for PostgreSQL kernel
  • PolarDB O: Domestic innovation database with Oracle-compatible kernel for local compliance requirements (Enterprise Edition subscription only)
• Extension Support Capability: Provides out-of-the-box installation for 576 available PG extensions for PG 14-18 on mainstream operating systems.
• Complete Functional Modules: Provides all functional modules:
  • Supabase: Reliably self-host production-grade open-source Firebase
  • Silo: Enterprise PB-level object storage planning and self-hosting
  • DuckDB: Provides comprehensive DuckDB support, and PostgreSQL + DuckDB OLAP extension plugin support
  • Kafka: Provides high-availability Kafka cluster deployment and monitoring
  • Kubernetes, VictoriaMetrics & VictoriaLogs
• Domestic Operating System Support: Provides domestic innovation OS support options (Enterprise Edition subscription only)
• Domestic ARM Architecture Support: Provides domestic ARM64 architecture support options (Enterprise Edition subscription only)
• China Mainland Mirror Repository: Smooth installation without VPN, providing domestic YUM/APT repository mirrors and DockerHub access proxy.
• Chinese Interface Support: Monitoring system Chinese interface support (Beta)

Payment Model

Pigsty subscription uses an annual payment model. After signing the contract, the one-year validity period is calculated from the contract date. If payment is made before the subscription contract expires, it is considered automatic renewal. Consecutive subscriptions have discounts. The first renewal (second year) enjoys a 95% discount, the second and subsequent renewals enjoy a 90% discount on subscription fees, and one-time subscriptions for three years or more enjoy an overall 85% discount.

After the annual subscription contract terminates, you can choose not to renew the subscription service. Pigsty will no longer provide software updates, technical support, and consulting services, but you can continue to use the already installed version of Pigsty Professional Edition software. If you subscribed to Pigsty professional services and choose not to renew, when re-subscribing you do not need to make up for the subscription fees during the interruption period, but all discounts and benefits will be reset.

Pigsty’s pricing strategy ensures value for money - you can immediately get top DBA’s database architecture construction solutions and management best practices, with their consulting support and comprehensive coverage; while the cost is highly competitive compared to hiring database experts full-time or using cloud databases. Here are market references for enterprise-level database professional service pricing:

• AWS RDS for PostgreSQL High Availability Edition: ¥1,160 ~ ¥1,582 / (vCPU·month), equivalent to 14K ~ 19K/year (per vCPU)
• Alibaba Cloud RDS for PostgreSQL High Availability Edition: ¥270 ~ ¥432 / (vCPU·month), equivalent to 3K ~ 5K/year (per vCPU)
• EDB PostgreSQL Cloud Database Enterprise Edition: $183.3 / (vCPU·month), equivalent to 16K/year (per vCPU)
• Fujitsu Enterprise PostgreSQL Kubernetes: $3200 / (Core·year), equivalent to 12K/year (per vCPU)
• Oracle Annual Service Fee: (Enterprise $47,500 + Rac $23,000) * 22% per year, equivalent to 28K/year (per vCPU)

The fair price for decent database professional services is 10,000 ~ 20,000 yuan / year, with the billing unit being vCPU, i.e., one CPU thread (1 Intel core = 2 vCPU threads). Pigsty provides top-tier PostgreSQL expert services in China and adopts a per-node billing model. On commonly seen high-core-count server nodes, it brings users an unparalleled cost reduction and efficiency improvement experience.

Pigsty Expert Services

In addition to Pigsty subscription, Pigsty also provides on-demand Pigsty x PostgreSQL expert services - industry-leading database experts available for consultation.

Contact Information

Please send an email to [email protected]. Users in mainland China are welcome to add WeChat ID RuohangFeng.

11 - FAQ

What is Pigsty, and what is it not?

Pigsty is a PostgreSQL database distribution, a local-first open-source RDS cloud database solution. Pigsty is not a Database Management System (DBMS), but rather a tool, distribution, solution, and best practice for managing DBMS.

Analogy: The database is the car, then the DBA is the driver, RDS is the taxi service, and Pigsty is the autonomous driving software.

What problem does Pigsty solve?

The ability to use databases well is extremely scarce: either hire database experts at high cost to self-build (hire drivers), or rent RDS from cloud vendors at sky-high prices (hail a taxi), but now you have a new option: Pigsty (autonomous driving). Pigsty helps users use databases well: allowing users to self-build higher-quality and more efficient local cloud database services at less than 1/10 the cost of RDS, without a DBA!

Who are Pigsty’s target users?

Pigsty has two typical target user groups. The foundation is medium to large companies building ultra-large-scale enterprise/production-grade PostgreSQL RDS / DBaaS services. Through extreme customizability, Pigsty can meet the most demanding database management needs and provide enterprise-level support and service guarantees.

At the same time, Pigsty also provides “out-of-the-box” PG RDS self-building solutions for individual developers, small and medium enterprises lacking DBA capabilities, and the open-source community.

Why can Pigsty help you use databases well?

Pigsty embodies the experience and best practices of top experts refined in the most complex and largest-scale client PostgreSQL scenarios, productized into replicable software: Solving extension installation, high availability, connection pooling, monitoring, backup and recovery, parameter optimization, IaC batch management, one-click installation, automated operations, and many other issues at once. Avoiding many pitfalls in advance and preventing repeated mistakes.

Why is Pigsty better than RDS?

Pigsty provides a feature set and infrastructure support far beyond RDS, including 576 extension plugins and 12+ kernel support. Pigsty provides a unique professional-grade monitoring system in the PG ecosystem, along with architectural best practices battle-tested in complex scenarios, simple and easy to use.

Moreover, forged in top-tier client scenarios like Tantan, Apple, and Alibaba, continuously nurtured with passion and love, its depth and maturity are incomparable to RDS’s one-size-fits-all approach.

Why is Pigsty cheaper than RDS?

Pigsty allows you to use 10 ¥/core·month pure hardware resources to run 400¥-1400¥/core·month RDS cloud databases, and save the DBA’s salary. Typically, the total cost of ownership (TCO) of a large-scale Pigsty deployment can be over 90% lower than RDS.

Pigsty can simultaneously reduce software licensing/services/labor costs. Self-building requires no additional staff, allowing you to spend costs where it matters most.

How does Pigsty help developers?

Pigsty integrates the most comprehensive extensions in the PG ecosystem (576), providing an All-in-PG solution: a single component replacing specialized components like Redis, Kafka, MySQL, ES, vector databases, OLAP / big data analytics.

Greatly improving R&D efficiency and agility while reducing complexity costs, and developers can achieve self-service management and autonomous DevOps with Pigsty’s support, without needing a DBA.

How does Pigsty help operations?

Pigsty’s self-healing high-availability architecture ensures hardware failures don’t need immediate handling, letting ops and DBAs sleep well; monitoring aids problem analysis and performance optimization; IaC enables automated management of ultra-large-scale clusters.

Operations can moonlight as DBAs with Pigsty’s support, while DBAs can skip the system building phase, saving significant work hours and focusing on high-value work, or relaxing, learning PG.

Who is the author of Pigsty?

Pigsty is primarily developed by Feng Ruohang alone, an open-source contributor, database expert, and evangelist who has focused on PostgreSQL for 10 years, formerly at Alibaba, Tantan, and Apple, a full-stack expert. Now the founder of a one-person company, providing professional consulting services.

He is also a tech KOL, the founder of the top WeChat database personal account “非法加冯” (Illegally Add Feng), with 60,000+ followers across all platforms.

What is Pigsty’s ecosystem position and influence?

Pigsty is the most influential Chinese open-source project in the global PostgreSQL ecosystem, with about 100,000 users, half from overseas. Pigsty is also one of the most active open-source projects in the PostgreSQL ecosystem, currently dominating in extension distribution and monitoring systems.

PGEXT.Cloud is a PostgreSQL extension repository maintained by Pigsty, with the world’s largest PostgreSQL extension distribution volume. It has become an upstream software supply chain for multiple international PostgreSQL vendors.

Pigsty is currently one of the major distributions in the PostgreSQL ecosystem and a challenger to cloud vendor RDS, now widely used in defense, government, healthcare, internet, finance, manufacturing, and other industries.

What scale of customers is Pigsty suitable for?

Pigsty originated from the need for ultra-large-scale PostgreSQL automated management but has been deeply optimized for ease of use. Individual developers and small-medium enterprises lacking professional DBA capabilities can also easily get started.

The largest deployment is 25K vCPU, 4.5 million QPS, 6+ years; the smallest deployment can run completely on a 1c1g VM for Demo / Devbox use.

What capabilities does Pigsty provide?

Pigsty focuses on integrating the PostgreSQL ecosystem and providing PostgreSQL best practices, but also supports a series of open-source software that works well with PostgreSQL. For example:

• Etcd, Redis, Silo, DuckDB, Prometheus
• FerretDB, Babelfish, IvorySQL, PolarDB, OrioleDB
• OpenHalo, Supabase, Greenplum, Dify, Odoo, …

What scenarios is Pigsty suitable for?

• Running large-scale PostgreSQL clusters for business
• Self-building RDS, object storage, cache, data warehouse, Supabase, …
• Self-building enterprise applications like Odoo, Dify, Wiki, GitLab
• Running monitoring infrastructure, monitoring existing databases and hosts
• Using multiple PG extensions in combination
• Dashboard development and interactive data application demos, data visualization, web building

Is Pigsty open source and free?

Pigsty is 100% open-source software + free software. Under the premise of complying with the open-source license, you can use it freely and for various commercial purposes.

We value software freedom. Pigsty uses the Apache-2.0 license. Please see the license for details.

Does Pigsty provide commercial support?

Pigsty software itself is open-source and free, and provides commercial subscriptions for all budgets, providing quality assurance for Pigsty & PostgreSQL. Subscriptions provide broader OS/PG/chip architecture support ranges, as well as expert consulting and support. Pigsty commercial subscriptions deliver industry-leading management/technical experience/solutions, helping you save valuable time, shouldering risks for you, and providing a safety net for difficult problems.

Does Pigsty support domestic innovation (信创)?

Pigsty software itself is not a database and is not subject to domestic innovation catalog restrictions, and already has multiple military use cases. However, the Pigsty open-source edition does not provide any form of domestic innovation support. Commercial subscription provides domestic innovation solutions in cooperation with Alibaba Cloud, supporting the use of PolarDB-O with domestic innovation qualifications (requires separate purchase) as the RDS kernel, capable of running on domestic innovation OS/chip environments.

Can Pigsty run as a multi-tenant DBaaS?

Pigsty uses the Apache-2.0 license. You may use it for DBaaS purposes under the license terms. For explicit commercial authorization, consider the Pigsty Enterprise subscription.

Can Pigsty’s Logo be rebranded as your own product?

When redistributing Pigsty, you must retain copyright notices, patent notices, trademark notices, and attribution notices from the original work, and attach prominent change descriptions in modified files while preserving the content of the LICENSE file. Under these premises, you can replace PIGSTY’s Logo and trademark, but you must not promote it as “your own original work.” We provide commercial licensing support for OEM and rebranding in the enterprise edition.

Pigsty’s Business Entity

Pigsty is a project invested by Miracle Plus S22. The original entity Panji Cloud Data (Beijing) Technology Co., Ltd. has been liquidated and divested of the Pigsty business.

Pigsty is currently independently operated and maintained by author Feng Ruohang. The business entities are:

• Hainan Zhuxia Cloud Data Co., Ltd. / 91460000MAE6L87B94
• Haikou Longhua Piji Data Center / 92460000MAG0XJ569B
• Haikou Longhua Yuehang Technology Center / 92460000MACCYGBQ1N

PIGSTY® and PGSTY® are registered trademarks of Haikou Longhua Yuehang Technology Center.

12 - Release Note

The latest Pigsty release is v4.5.0.

v4.5.0

Pigsty v4.5.0 is a feature release focused on new pilot modules, replaceable data services, cluster-identity-aware orchestration, observability, and the software supply chain. It introduces Kafka KRaft and MySQL 8.4 modules, adds Valkey to REDIS, converges the MINIO module on Silo, and expands the packaged extension catalog from 531 to 575 extensions. Released on 2026-08-15. See the GitHub release and the complete source comparison at v4.4.0...v4.5.0.

Highlights

• 575 extensions: Compared with v4.4.0’s 531 entries, the current catalog adds 46 and removes 2, for a net gain of 44. It now contains 575 extensions across 406 non-contrib package families, with RPM/DEB coverage tracked per platform.
• Kafka KRaft module: Adds native Pigsty orchestration for Kafka, with multi-cluster support, dynamic member enrollment and retirement, SCRAM/TLS, secure credential rotation, monitoring metrics, and Grafana dashboards.
• MySQL 8.4 module: Adds standalone and three-node InnoDB Cluster deployments, MySQL Router, XtraBackup, user and database provisioning, monitoring and alerting, and idempotent reconciliation.
• Valkey and Silo: REDIS adds redis_type: valkey; the final MINIO source accepts only minio_type: silo. The RustFS integration developed during this cycle was fully withdrawn before the candidate baseline.
• 51 standalone configuration templates: Adds demo/kafka, demo/mysql, and the eight-node ha/octo simulation template to the 48 standalone templates in v4.4.0. The compatibility symlink conf/app/supa.yml → ../supabase.yml remains available.
• Safer cluster-identity orchestration: PGSQL, REDIS, MINIO, KAFKA, and MYSQL initialization playbooks, plus every corresponding removal playbook except mysql-rm.yml, skip unrelated hosts by explicit cluster identity. mysql-rm.yml instead fails closed for any wrongly selected host. etcd delegation and DBSU key exchange now use actual cluster members as well.
• Observability and supply chain: Re-exports Grafana dashboards to Dashboard API v2, moves MinIO/Silo collection to Metrics V3, and generates local repositories atomically with SOW instead of synthetic ModuleMD metadata.
• Kernel and toolchain updates: Completes pgBackRest support for PostgreSQL 19 beta2, enables cluster mode for Percona PostgreSQL TDE, fixes IvorySQL initialization and WAL compression, and refreshes extension package maps, exporters, and build tooling.

New Modules and Data Services

• The Kafka module uses node state as the source of truth for dynamic KRaft orchestration. It manages one or more clusters in one inventory and also supports an unbounded kafka.yml run; partial --limit selections are rejected. The destructive kafka-rm.yml instead requires a non-empty -l/--limit and validates a safe absolute data path plus surviving broker/controller anchors before any partial retirement can stop services. Nodes retain authoritative manifests and secrets, with dynamic controller joins, broker admission, member retirement, three-step dead-node replacement, SCRAM-SHA-512/TLS, credential and certificate rotation, and self-tested partition health gates.
• The MySQL pilot module targets a fixed MySQL 8.4 LTS platform and accepts either a standalone node or a three-node InnoDB Cluster. It includes MySQL Shell and Router, scheduled full XtraBackup backups, TLS, account and database provisioning, primary-key policy checks, conservative member removal, and idempotent reconciliation.
• The REDIS module retains redis as its default engine and can deploy Valkey with redis_type: valkey. Service units now use Type=notify with a 1,800-second startup timeout, plus stronger topology validation, password handling, tag-scoped removal semantics, and rebuild protection.
• The MINIO module now deploys Silo and only Silo. minio_type remains an extension point, but silo is the sole accepted value in this release. Startup checks the systemd Invocation ID and ActiveState=active for the current restart, waits about 600 seconds by default, and then runs Silo’s cluster health check. The Infra package line adds silo and mcli while preserving the S3/Admin APIs, /minio/* routes, MINIO_* environment variables, and disk format.
• Object-storage topology is grouped by minio_cluster; its inventory group name may differ, and one inventory may declare multiple object-storage clusters. Use distinct minio_alias, minio_domain, and minio_endpoint values for each to avoid overwriting shared client aliases on INFRA nodes. demo/minio now selects Silo explicitly and trims its local repository to the infra,node modules.
• The standalone FERRET module is replaced by PostgreSQL Mongo mode and the FerretDB Docker APP. PostgreSQL provides the DocumentDB data layer, while Docker Compose provides the FerretDB protocol layer.

Orchestration, Security, and Tooling

• deploy.yml, slim.yml, and the PGSQL, REDIS, MINIO, KAFKA, and MYSQL initialization playbooks now skip unrelated hosts according to the corresponding *_cluster identity. The PGSQL, REDIS, MINIO, and KAFKA removal playbooks do the same. MySQL removal is intentionally different: mysql-rm.yml does not skip hosts without identity and instead fails closed in mysql_rm_check. Every host that enters a role still receives an internal identity check.
• PGSQL configuration, PITR, and removal workflows delegate only when the canonical etcd group exists and has at least one member; they no longer silently fall back to localhost when no etcd target exists. DBSU SSH keys are exchanged through the actual pg_cluster_members, correctly covering cross-inventory-group topologies such as Citus.
• PGSQL PITR and removal now delete only the etcd subtree bounded by /<cluster>/, avoiding adjacent clusters whose names share a prefix. The initial pgBackRest marker /etc/pgbackrest/initial.done is written only after the backup command succeeds.
• HAProxy uses the fixed /etc/haproxy/haproxy.cfg and /etc/haproxy/conf.d layout, upstream master-worker mode, a master socket, and Type=notify. dnsmasq now binds dynamically, answers private reverse lookups locally, and handles node addresses added after INFRA initialization.
• Rendered systemd units managed by Pigsty are consistently placed under /etc/systemd/system; permissions on sensitive configuration and privileged files are tightened further. Removal workflows stop services before entering the data-cleanup phase.
• The REPO and CACHE roles now use sow create --pigsty to atomically generate RPM/APT metadata and the SHA-256 repo_complete marker, and no longer generate synthetic ModuleMD metadata. pg_id also compares cluster size as an explicit integer for older Ansible releases.
• RPM exporter package names move from underscores to hyphens, for example node_exporter → node-exporter. Debian repository naming and PGDG YUM extension package mappings are corrected as well.
• Tuned profiles now use the OS-specific directory: /etc/tuned/profiles on EL 10, Debian 13, and Ubuntu 26, and /etc/tuned on EL 8/9, Debian 12, and Ubuntu 22/24. Debian/Ubuntu package installation also suppresses premature starts of Silo, Redis/Valkey, and legacy log services.
• China-region repository routing receives a systematic refresh. OS, Docker, Grafana, Percona, supported MongoDB APT, and uv/PyPI paths prefer Tencent Cloud; EL and Docker entries retain Huawei Cloud and Aliyun fallbacks where appropriate. MySQL and Kubernetes use USTC mirrors, and ClickHouse uses Huawei Cloud. MongoDB RPM no longer advertises an unavailable China-region alternative. The final per-platform choices remain defined in roles/node_id/vars/<os>.<arch>.yml.
• The Docker image moves to Debian 13.6 and Pigsty v4.5.0. Vagrant enforces a 32 GiB root disk, accepts pinned box versions, and adds the eight-node ha/octo lab. docker/Makefile fixes its data directory at ./data, and make purge deletes that directory directly.
• GitHub Actions for checkout, CodeQL, Docker build/login, and Cosign are upgraded in one batch. Release, bootstrap, install, and validation scripts also tighten file and argument handling. Release archives now derive top-level pigsty.yml from conf/meta.yml, include the Kafka/MySQL playbooks, and drop the legacy Mongo playbook.
• The release-signing workflow must be dispatched from main and signs only the pigsty-<tag>.tgz source archive; multi-gigabyte offline bundles are outside that workflow. The package-build bootstrap matrix now matches conf/build/oss.yml: EL 9/10, Debian 12/13, and Ubuntu 22/24/26.

Observability

• Re-exports the dashboard set through the Pig/Grafana tooling to Dashboard API v2. Adds four Kafka and five MySQL dashboards, and refreshes links, variables, and layouts across Node, PostgreSQL, Redis, and Infra dashboards.
• Migrates the MinIO/Silo Overview and Instance dashboards to Metrics V3. Victoria scrapes the /minio/metrics/v3 root endpoint and drops high-cardinality samples carrying a non-empty bucket label.
• Updates the pg_exporter configuration to 1.4.0 and fixes duplicate time series from the 1.4.1 pg_subrel query. For PG19 it adds pg_sub_19, pg_recovery_state, pg_wal_19, pg_lock_stat, and pg_vacuum_score; PG10+ gains the pg_xact_age transaction-age histogram, and replication-slot idle_timeout plus WAL Receiver connecting state encoding are covered. Kafka JMX/protocol exporters and the MySQL exporter also join the standard target and alert pipelines.

PostgreSQL Kernels and Extension Packages

• PostgreSQL 19 beta3 templates now include pgBackRest packages and backup support.

• All four standard Patroni templates add the PostgreSQL 18.6 logical-decoding allowlist output_plugin_libraries: 'pgoutput, test_decoding, wal2json'; Patroni filters it on older PostgreSQL versions that do not support the setting.

• Percona PostgreSQL 18 TDE now uses cluster mode and retains Pigsty-prefixed packages to avoid conflicts with native PostgreSQL packages.

• IvorySQL now initializes its default database correctly and enables compatible WAL compression in workload templates.

• The PostgreSQL fact loader, per-platform package_map, and default extension groups are refreshed to fill package gaps and correct PGDG/YUM naming. Comparing names between the v4.4.0 PIG v1.5.1 catalog and the current catalog yields 46 additions and 2 removals:

  • 32 new primary extensions: argm, cat_tools, cron_utils, fbsql, oidc_validator, online_advisor, pg_cjk_parser, pg_column_tetris, pg_describe, pg_disorder, pg_fts, pg_jieba, pg_kpart, pg_lake, pg_local_cache, pg_mentat, pg_oidc_validator, pg_policy, pg_roast, pg_tiktoken_c, pg_turbovec, pg_vault_tde, pgcontext, pgfr_record, pgmemento, pgmonitor, pgsqlmock, pgwasm, plruby, plx, postbis, and qdgc.
  • 13 child extensions from those package families: hstore_plruby, jsonb_plruby, ltree_plruby, pg_extension_base, pg_extension_updater, pg_lake_copy, pg_lake_engine, pg_lake_iceberg, pg_lake_table, pg_map, pgcontext_pgvector, pgfr_analyze, and qdgc_postgis.
  • One new PGDG extension, pg_statviz. Its package is hidden from the default install group, but the extension remains in the online catalog.
  • Two catalog removals: pg_analytics and spat. The total therefore rises from 531 to 575, a net gain of 44.

• Cumulative notable upgrades include citus 14.2.0, pg_search 0.25.2, timescaledb 2.29.1, vector 0.8.6, documentdb 0.114, pg_partman 5.5.0, pgmnemo 0.16.1, plpgsql_check 2.10.4, provsql 1.12.0, and pgbson 2.1.0, plus a broad pgrx 0.19.1 rebuild.

• Full build records and platform differences appear in the merged table below and the original RPM changelog and DEB changelog. New catalog entries include pg_local_cache and pg_policy. Changelog dates identify package batches and should not be equated one-for-one with the current CSV mtime.

• pg_statviz is excluded only from the default install group in db/reload.sql; its detail page, platform coverage, and package-name differences remain in the online catalog.

Extension Package Update Log

The table below merges the RPM changelog and DEB changelog after v4.4.0, with 231 rows aligned by batch and extension name. Records that are identical in RPM and DEB are merged; version or note differences are shown separately. An unchanged version still indicates a rebuild, package-name change, license-metadata change, or platform-coverage change.

The first extension batch after July 10 is July 24. That original batch covers July 7–24 without per-item dates, so it is included in full to avoid omitting post-release pgrx rebuilds and package-matrix fixes. “RPM only” or “DEB only” means only that the other changelog has no same-batch row for that extension.

Infrastructure Package Update Log

The following table contains all 144 Infra log records after v4.4.0, from 2026-07-16 through the latest 2026-08-12 batch. It also includes the ferretdb2 rebuild and RPM exporter package-name migration recorded in the changelog prose. Consecutive upgrades of the same package are retained as separate batch entries.

Build, download, and verification status follows the wording of the original log; it does not establish repository indexing, signing, synchronization, or offline-bundle acceptance. See the Infra changelog for full context.

Compatibility Changes and Upgrade Notes

• Existing FERRET deployments should remove the legacy ferretdb systemd service and redeploy the protocol layer with docker.yml and app.yml. The old mongo.yml playbook, mongo_* parameters, scrape job, and dedicated dashboard are no longer provided.
• Pigsty no longer renders /etc/default/haproxy for the HAProxy unit, although the unit can read it when present. Use only EXTRAOPTS for process arguments, not OPTIONS. Any EXTRAOPTS override must retain -S /run/haproxy-master.sock and must not include -f.
• New module playbooks require target hosts to define the corresponding pg_cluster, redis_cluster, minio_cluster, kafka_cluster, or mysql_cluster explicitly. Custom inventories that relied on fixed group names without cluster identity variables must add those identities first.
• The MINIO role now accepts only minio_type: silo; minio and rustfs fail during identity validation. Silo retains MinIO protocol and on-disk compatibility, but the package, binary, and systemd service names change. Back up existing object storage and validate in-place compatibility and rollback before upgrading; do not treat package replacement as a migration that has already passed acceptance.
• Valkey remains opt-in. redis_type: valkey installs valkey-server and valkey-cli, while configuration paths, service names, monitoring jobs, and other module-facing interfaces remain under redis for compatibility.
• Pigsty’s core REPO/CACHE roles require SOW and use sow create --pigsty to generate local repositories. Older offline bundles or local repositories without SOW 0.3.0 must first install or refresh it from the Pigsty INFRA repository. pig repo create is a separate CLI path whose fallback behavior depends on its own version.
• MySQL mysql_databases entries accept only name, encoding, and collate, and databases are created with DEFAULT ENCRYPTION='N'. mysql_parameters cannot use loose_, skip_, disable_, or enable_ prefixes to bypass platform-owned, replication, or TLS option protection.
• Custom RPM repositories and external automation that still reference underscore names such as node_exporter or redis_exporter must move to the hyphenated node-exporter and redis-exporter package names.
• docker/Makefile no longer accepts DATA to redirect the cleanup target. make purge deletes repository-local ./data immediately without a countdown; preserve any required data first.
• KAFKA and MYSQL remain pilot modules. Kafka clients must resolve and reach each broker directly rather than putting the data plane behind HAProxy, a VIP, or an L4 load balancer. MySQL currently accepts exactly one or three members.

All 14 validated offline artifacts are published on GitHub for this release, one per recommended OS version and architecture, alongside a checksums manifest and a detached PGP signature (.asc) for every file.

Checksums

e042059379bdfae8f774022b89e8d1e3  pigsty-pkg-v4.5.0.el9.aarch64.tgz
997e812a433a6b969b976fad2c023a1f  pigsty-pkg-v4.5.0.el9.x86_64.tgz
1e1045db965282d564680534bd7d72e2  pigsty-pkg-v4.5.0.el10.aarch64.tgz
9a53f1e85cbb2d4f85969a6112ae4b05  pigsty-pkg-v4.5.0.el10.x86_64.tgz
b7501783c90311176f21bdd35390c746  pigsty-pkg-v4.5.0.d12.aarch64.tgz
f3ecaa449a0bf8e0f01907f83831e74a  pigsty-pkg-v4.5.0.d12.x86_64.tgz
863165dba76b044ed8615d6743710005  pigsty-pkg-v4.5.0.d13.aarch64.tgz
d86655361ccad7aa95a345a82bb37d10  pigsty-pkg-v4.5.0.d13.x86_64.tgz
017f2d7931eb644d2d0fa2f71930134e  pigsty-pkg-v4.5.0.u26.aarch64.tgz
61451ee610134423ff08f1a69dfced33  pigsty-pkg-v4.5.0.u26.x86_64.tgz
5d9cfc52a25545b56e73e94ab5b5e175  pigsty-pkg-v4.5.0.u24.aarch64.tgz
dba0eef49899509d1524b3a1c37d0ddc  pigsty-pkg-v4.5.0.u24.x86_64.tgz
5564841c7c099489708cd1fe49ffa1b9  pigsty-pkg-v4.5.0.u22.aarch64.tgz
dc52b6cee50cf6226e23b065e5aa8395  pigsty-pkg-v4.5.0.u22.x86_64.tgz
afb5cd77903613cb945bd519e4059c76  pigsty-v4.5.0.tgz

v4.4.0

Pigsty v4.4.0 is a maintenance release centered on PostgreSQL 18.4, PostgreSQL 19 beta readiness, 531 extensions, refreshed kernel variants, and broader platform coverage.

Released on 2026-07-10. See the GitHub release and all changes since v4.3.0.

Highlights

• PostgreSQL 18.4 / 19 beta: PostgreSQL 18.4 is now the production default, with a minimal PostgreSQL 19 beta template for evaluation.
• 531 extensions and refreshed kernels: The catalog adds 21 extensions and updates major PostgreSQL variants across the supported platform matrix.
• Safer operations with Pig 1.5.1: New clone, fork, and PITR workflows arrive alongside automatic VIP discovery, Zstandard pgBackRest compression, and dedicated Patroni log collection.
• Security, applications, and tooling: Secret handling and repository automation are hardened, with new app templates, a redesigned portal, and optional Codex support.
• Platform validation: All 14 offline deployment tests pass across seven OS baselines on both x86_64 and aarch64.
• Offline artifacts: The Community Edition publishes six dual-architecture offline packages for Debian 13, EL 10, and Ubuntu 24.04 on GitHub. Prebuilt packages for the other validated baselines are available with the Professional Edition.

Upgrade Notes

• Generated pgBackRest configurations now use compress-type=zst; preserve intentional local overrides before re-rendering them. #744
• Patroni logs now use /pg/log/patroni and job=patroni; update custom log queries and alert rules that use the old syslog selector.
• VIP interfaces now default to auto, dnsmasq records move to /etc/dnsmasq.d/pigsty, and Pigsty manages /etc/default/haproxy; preserve explicit network overrides where needed.
• The default etcd backend quota drops from 16 GiB to 8 GiB; check existing backend usage before applying the new configuration.
• pig automation must use -y/--yes for destructive commands, while pig pb restore and pig pitr require one explicit recovery target. See the pig v1.5 notes.
• Supabase analytics moves to the _supabase database and _analytics schema; existing deployments should create them before switching stacks.

Security and Operations

• The pg-pitr wrapper adds safer target selection, timelines, dry runs, and stronger checks against unsafe recovery targets.
• Application secrets are hidden from Ansible output, generated .env files use mode 0600, and Grafana no longer prints the administrator password.
• The dbsu sudo policy gains controlled journal access, while the repository adds a security policy, CodeQL, Dependabot, pinned actions, and release-signing automation.

Applications and Tooling

• Added Immich, Maybe, and JumpServer templates; refreshed Supabase, Dify, InsForge, Registry, Jupyter, Kong, Odoo, Teable, Mattermost, and related launch helpers.
• Rebuilt the bilingual infrastructure portal and added opt-in Codex CLI support to the experimental VIBE module; Claude Code remains its default managed coding agent.
• Removed the legacy FerretDB Compose template; the FERRET module remains available.

Bug Fixes

• Fixed EL10 PostgreSQL/libpq provider conflicts, EPEL path handling, and PGDG minor-version repository rules. #752
• Reused an existing /www directory during bootstrap and fixed Redis Sentinel HA password rendering. #753 #748
• Corrected RPM naming and package groups for pg_http, pg_gzip, apache-age, and odbc_fdw. #750
• Prevented unexpected service starts during Debian and Ubuntu package installation, and improved EL9 aarch64 Patroni package handling.
• Fixed VirtualBox private-network routing and default NIC selection.
• Fixed shell portability and Vector log lifecycle issues, along with PG19 io_workers, Teable HBA, and application runtime defaults.

PostgreSQL and Extension Package Changes

The release adds 21 extensions, updates the PostgreSQL 18.4 package graph, introduces the PostgreSQL 19 beta template, and refreshes major kernel variants. Versions below are verified against final repository metadata and, where bundled, the v4.4.0 artifacts; PG major ranges describe catalog and repository coverage.

PostgreSQL RPM changes · PostgreSQL DEB changes · Infrastructure package changes

Infrastructure Package Changes