TL;DR
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
PostgreSQL 19 is likely to arrive in November rather than within the project’s usual mid-September to mid-October release window, according to developer Robert Haas. Haas attributes the delay and several late feature reverts in part to more bugs being found with large language models, while noting that people still need to review and approve fixes.
PostgreSQL 19 is likely to be released in November, later than the project’s usual mid-September to mid-October window, developer Robert Haas said. In a blog post, Haas linked the delay to a rise in bugs reported with help from large language models (LLMs), which has increased the work required to review fixes and assess release features.
Haas said the project saw three major features reverted late in the beta period, between August 27 and September 15: UPDATE/DELETE FOR PORTION OF, SQL Property Graph Queries (SQL/PGQ), and ALTER TABLE … MERGE PARTITION/SPLIT PARTITION. Feature removals after the annual April 8 feature freeze are not unheard of, he wrote, but several major reverts so late in the process are unusual.
Haas compared commit counts over the same dates in consecutive years. Between April 8 and September 25, 2025, PostgreSQL recorded 724 commits to REL_18_STABLE; in the corresponding 2026 period, REL_19_STABLE had 1,169. He also reported that commits to the prior stable branch rose from 248 to 433, comparing REL_17_STABLE in 2025 with REL_18_STABLE in 2026. Those counts measure commits, not bugs, and do not by themselves establish their cause.
The master-branch figures moved differently: Haas counted 1,037 commits in the 2025 period and 1,443 in the equivalent 2026 period, while saying that master-branch activity between feature freeze and September 25 had declined slightly year over year. He argued that increased back-patching reflects more defects being found in released branches, rather than a clear rise in mistakes by developers.
Why More Bug Reports Delay Release
The delay reflects a tension between shipping on a predictable schedule and releasing only after the project has had time to review changes and address problems. More bug reports can improve software quality, but each report takes developer and reviewer time, and fixes proposed with AI assistance still need human scrutiny.
Haas said attention to bugs and security issues in older release branches also drew time away from reviewing what should remain in PostgreSQL 19. The immediate consequence is a later release and the removal of features; the longer-term issue he identifies is whether the project can keep up with defects that new tools may surface faster than maintainers can resolve them.
Products Worth Considering
Bose SoundLink Flex Bluetooth Speaker (2nd Gen), Black
Apple Watch Series 12 GPS Smartwatch, 46 mm
Bose QuietComfort Ultra Wireless Noise-Cancelling Headphones with Spatial Audio, Black
As an affiliate, we earn on qualifying purchases.
PostgreSQL’s Usual Release Window
Haas described the project’s major-release timing over the past decade as relatively consistent: releases have generally arrived between mid-September and mid-October. He said PostgreSQL 19 is expected to miss that window, with November the most likely timing, though a later release remains possible.
The project’s annual feature freeze occurs on April 8, after which the release process focuses on testing, fixing problems, and deciding which changes are ready to ship. Haas noted that feature reverts immediately after the freeze are not especially unusual; his concern was the number and late timing of major reverts this year.
The explanation comes from Haas’s account of project activity, published on his blog and originally appearing on Neon’s blog. His explanation that LLM-assisted reports are a major cause is an assessment by a PostgreSQL developer, rather than a formal project finding establishing the cause of the delay.
Release Date and AI’s Exact Role
No exact PostgreSQL 19 release date is given. Haas said November is the likely timing but that a further postponement cannot be ruled out. The source does not provide a confirmed revised schedule from the PostgreSQL project.
The commit totals show more activity in stable branches, but they do not identify how many changes were prompted by LLM-assisted reports, how many reports were valid, or how much review time each one required. Haas presents LLM-assisted bug discovery as the likely main explanation, while acknowledging that other factors, including possible changes in developer mistakes, cannot be excluded by the figures alone.
Testing Before PostgreSQL 19 Ships
The project is expected to continue reviewing fixes and deciding which changes are safe to include before setting or confirming a release date. Haas’s account indicates that maintainers may need to balance PostgreSQL 19 work against bug and security fixes for supported older branches.
Readers should watch for an official release schedule and further project updates on the status of the reverted features. Until those arrive, November remains an estimate, not a confirmed launch date, and it is unclear whether any of the three removed features could return in a later release.
Key Questions
When is PostgreSQL 19 expected to be released?
Developer Robert Haas said November 2026 is the most likely timing, later than the project’s typical mid-September to mid-October window. He also said a further delay is possible; no exact date was provided.
Why does Haas say the release is late?
Haas attributes the delay mainly to more bugs being reported after discovery with large language models, alongside late feature decisions and work on bugs and security issues in older branches. This is his assessment, not a formally established project finding.
Which PostgreSQL 19 features were reverted?
Haas named UPDATE/DELETE FOR PORTION OF, SQL Property Graph Queries (SQL/PGQ), and ALTER TABLE … MERGE PARTITION/SPLIT PARTITION. He said these reverts occurred between August 27 and September 15.
Do the commit numbers count bugs found?
No. The figures Haas provided count commits to PostgreSQL branches over matching periods in 2025 and 2026. They do not directly count bug reports or prove what caused the increase in back-patched changes.
Source: rhaas.blogspot.com
