Version Overview

On March 12, 2026, the IvorySQL community officially released IvorySQL 5.3.

Built on PostgreSQL 18.3, this release keeps pace with the latest stable PostgreSQL capabilities while further strengthening cloud-native deployment, ecosystem extensibility, and overall user experience. IvorySQL 5.3 also introduces an online trial environment so users can explore it directly in the browser without local installation.

For the complete update list, visit the official documentation site at docs.ivorysql.org.

Enhanced Features from PostgreSQL 18.3

  • Fixed failures that could occur after replaying multixid truncation WAL records generated by older minor versions.
  • Avoided incorrect invalid-encoding errors when applying substring() to toasted data.
  • Addressed an omission in the fix for CVE-2026-2007.
  • Corrected the volatility markings of json_strip_nulls() and jsonb_strip_nulls().
  • Fixed the computation of outer joins that can produce null values in LATERAL UNION ALL subquery outputs.

For more details, see the official PostgreSQL 18.3 Release Notes.

Enhanced Features from PostgreSQL 18.2

  • Guarded against unexpected dimensions in oidvector and int2vector values.
  • Improved selectivity estimators to avoid attaching them to operators that accept unexpected data types.
  • Fixed a buffer overflow in PGP decryption functions in the contrib/pgcrypto module.
  • Fixed insufficient validation of multibyte character lengths.
  • Enhanced the contrib/pg_trgm module to cope with changes in string lowercasing behavior.

For more details, see the official PostgreSQL 18.2 Release Notes.

New Features of IvorySQL 5.3

  • Upgraded to the PG 18.3 kernel.
  • Online Trial: a browser-based interactive environment that lets users explore and test IvorySQL 5.3 without local installation.
  • Full-platform installation package support across X86, ARM, MIPS, and LoongArch architectures.
  • Containerized deployment support with one-click delivery of standalone or high-availability IvorySQL 5.3 clusters through Docker Compose, Podman, Swarm, Helm, and Operator.
  • IvorySQL Cloud 5.3 released with a fully managed visual control plane for subscription management, lifecycle operations orchestration, and ecosystem service integration.
  • Support for 10 additional PostgreSQL extensions: pg_cron, pgAudit, PostGIS, pgRouting, PGroonga, ddlx, pgsql-http, system_stats, plpgsql_check, and pgvector.

Fixed Issues

  • Cleaned up compilation warnings during the build process.
  • Fixed the raw_expression_tree_walker handling issue for ColumnRefOrFuncCall.
  • Fixed failures when upgrading from PostgreSQL to IvorySQL.
  • Improved PL/iSQL error messages when ivorysql.compatible_mode = 'pg'.
  • Fixed an out-of-bounds read in token_is_col_id() within liboracle_parser.c.
  • Fixed the issue where psql tab completion incorrectly displayed INVISIBLEINCREMENT.
  • Fixed variable reset issues in nested function expressions.
  • Fixed type coercion failures when packages and subprograms mixed positional and named arguments.
  • Fixed a potential segmentation fault when package procedures initialized local variables from parameters.
  • Fixed regression test failures caused by cross-year execution.

Source Code Repositories

The development and maintenance of IvorySQL are mainly distributed across the following four repositories:

Contributors

The following individuals, listed in alphabetical order, contributed to this release as code authors, committers, reviewers, testers, or issue reporters:

Cédric Villemain, Fawei Zhao, Ge Sui, Grant Zhou, Oreo Yang, Pierre Forstmann, Rophy Tsai, Shuntian Jiao, Steven Niu, Xiangyu Liang, Xiaohui Liu, Xinjie LYU, Xueyu Gao, Yasir Hussain Shah, Yuan Li, Zheng Tao, Zhenhao Pan, Zhe Zhang, Zhuoyan Shi.

Thanks to all the contributors!

Try It Out Now

The complete changelog is available at https://github.com/IvorySQL/IvorySQL/compare/IvorySQL_5.1...IvorySQL_5.3.

We welcome everyone in the community to try IvorySQL 5.3. If you encounter any issues, please submit feedback through GitHub Issues. We will send IvorySQL merchandise as a thank-you gift.