Sammlung von Newsfeeds

Parallel apply workers speed up logical replication of large transactions—but only if you have enough of them.

Neither of these parameters means what its name says. max_parallel_workers_per_gather is the most workers the planner is allowed to request for one plan node, and the planner usually requests fewer. max_parallel_workers is documented as a limit for the whole cluster, and it is enforced by compari…

I did it! The most important thing I anticipated after leaving DRW was having more time for important work, and one of the top items on my to-do list was the pg_acm update.

I already knew about two dozen bugs I needed to fix; several new functions people asked about. In addition, I had several conceptual changes in mind. Do you know how disturbing it is when you know you need to do something, and you can’t focus on that “something” because other things, less important but more urgent, keep popping up?

max_parallel_maintenance_workers is a ceiling, and most of the time something lower gets there first. It caps the number of parallel workers a single utility command may start. The default is 2, the context is user, and the range is 0 to 1024, with 0 turning parallel maintenance off. Raising it i…

I used AI tools to size an OLTP workload on an EC2 system, with DBT-5, a TPC-E-like fair-use implementation. I provided a systematic and mechanical plan for running a series of tests to determine what the appropriate scale factor is on a system.

Never set your PostgreSQL locale to zh_CN — that old rule expires with PostgreSQL 19. The Chinese message catalogs went seven years untouched and nearly got dropped from the release. A bone nobody could chew through in seven years, finished in seven days.

The future of Postgres is bright. So bright in fact, that I spent almost a dozen posts expounding on the upcoming features it would bring, with veritable stars in my eyes. Unfortunately, while I was out counting my chickens, it would seem some of these exciting new features failed to hatch.How, and perhaps more importantly why that happened, deserves some investigation.

postgres.scot is live

Postgres events in Scotland now have a permanent address: postgres.scot

Nobody asked for max_notify_queue_pages. It exists because a limit that used to enforce itself stopped doing so, and something had to take its place. LISTEN/NOTIFY runs on a single queue for the whole cluster, stored as SLRU pages under pg_notify/ in the data directory. Through PostgreSQL 16 that…

Two client connections, opened one after the other through PgBouncer transaction mode, asked the same database whose rows they were allowed to see, and the second one got the first one's answer. It had set nothing, it had never met the first caller, and the rows it read belonged to that caller's tenant. Any runbook that keeps a tenant key in a session variable sits one pooler away from this, and since 28 July the MCP protocol carries no session of its own, so a tool call from an agent arrives in this shape by default.

SourceDocsProduct page2026-09-17pgColumnar 1.0-alpha4 release notesRelease date: 2026-09-17 Previous release: 1.0-alpha3 (2026-09-02)pgColumnar is a columnar table access method for PostgreSQL. This is the fourth alpha. Its theme is layout and skipping. A table can now be laid out on the Hilbert curve, which keeps neighbouring keys closer together than Z-order does. A star-schema join now skips fact-table groups and …

Role-level timeouts and read-only flags sound safe until you realize the session can change them in one statement.

Ask a database vendor about on-prem and you'll usually get a one-line answer: 'we're open source, you can just self-host it.' For a hobby project, fine. For an enterprise, that line rarely survives contact with production. And enterprises want on-prem more than ever right now, largely because of AI. Running open models on hardware you own is far cheaper than renting inference by the token, and in a regulated industry, keeping data where you control it isn't optional.

Last week, I spent three days in the Netherlands and gave two talks at two conferences: a lightning talk at PGDay Lowlands in Utrecht on Thursday, September 10, and a session at Percona Live in Amsterdam on Friday, September 11\. In this blog post, I’m going to share my notes from both.

As a SQL Server developer and DBA learning Postgres, it’s easy to expect that the information you need for meaningful query and performance tuning will be readily available. For years (decades, maybe) you’ve learned the DMVs, set up Extended Events sessions, relied on Query Store, and regularly run Ola Hallengren’s maintenance scripts and Brent Ozar’s First Responder Kit.

As a SQL Server developer and DBA learning Postgres, it’s easy to expect that the information you need for meaningful query and performance tuning will be readily available. For years (decades, maybe) you’ve learned the DMVs, set up Extended Events sessions, relied on Query Store, and regularly run Ola Hallengren’s maintenance scripts and Brent Ozar’s First Responder Kit.

In PostgreSQL, every tuple starts with 23-byte header, and the first eight bytes are two transaction IDs. t_xmin for the transaction that created the row and t_xmax for the one that deleted or updated it. That is the visibility story covered in PostgreSQL MVCC, Byte by Byte. For now we have discussed t_xmax acting as the delete marker.

When a logical replication worker pool runs dry, subscriptions don't fail—they just stop making progress, filling your logs with warnings every five seconds…

Postgres 19 won't make the expected release date. PostgreSQL has shipped its major version every fall for the last several years. But this year, the code is in a heavy review cycle, major features have been reverted during beta, and many others are under heavy revision. Beta 4 is scheduled for Sept. 24, 2026. A release of Postgres 19 is certainly delayed by weeks and maybe even months.

Version 19: 71 commits on REL_19_STABLE. Version 20: 46 commits on master. Reverted out of 19: 74 commits, in 6 reverts. Six features tracked, 2 still shipping. Window closed at d8408e8d682.Summary71 commits landed on the 19 branch in nine days. The six reverts in the same window name 74 commits between them.Six reverts, newest first.Online data checksum transitions. September 16, Daniel Gustafsson.UPDATE …