News for hackers. Stuff that matters.

A comment on Explaining the Postgres Meme. Parent comment

Re:Explaining the Postgres Meme by alawrence on Thursday September 07, 2023 @11:16AM (#37420194)
> so its conservative approach of only acquiring the lock if there's no contention is completely thwarted.

How is contention on the standbys communicated back to the writer? I'm confused about how the writer is interrupted mid-cycle if only the readers are experiencing lock contention.

[ Parent | on HN ]
Re:Explaining the Postgres Meme by phamilton on Thursday September 07, 2023 @03:45PM (#37424565)
The writer also has traffic that can attempt to acquire the locks. Reads and writes (though our highest volume read traffic is reader only).

Contention from the standbys do not get communicated to the writer.

[ Parent | on HN ]