Re: Our poll() based WaitLatch implementation is broken - Mailing list pgsql-hackers

From Peter Geoghegan
Subject Re: Our poll() based WaitLatch implementation is broken
Date
Msg-id CAEYLb_UpKveDsjeXoweCQuOZH7Kk4O9QbaHTwbmBMqiR9UdP7w@mail.gmail.com
Whole thread Raw
In response to Our poll() based WaitLatch implementation is broken  (Peter Geoghegan <peter@2ndquadrant.com>)
List pgsql-hackers
On 15 January 2012 07:26, Peter Geoghegan <peter@2ndquadrant.com> wrote:
> Build Postgres master, on Linux or another platform that will use the
> poll() implementation rather than the older select(). Send the
> Postmaster SIGKILL. Observe that the WAL Writer lives on, representing
> a denial of service as it stays attached to shared memory, busy
> waiting (evident from the fact that it quickly leaks memory).

Correction: The CPU usage % approaches 100.

--
Peter Geoghegan       http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training and Services


pgsql-hackers by date:

Previous
From: Greg Smith
Date:
Subject: Vacuum rate limit in KBps
Next
From: Greg Smith
Date:
Subject: Re: automating CF submissions (was xlog location arithmetic)