Skip site navigation (1) Skip section navigation (2)

Swap Space and vm.oom_kill_allocating_task

From: Volkan YAZICI <yazicivo(at)ttmail(dot)com>
To: pgsql-general(at)postgresql(dot)org
Subject: Swap Space and vm.oom_kill_allocating_task
Date: 2008-05-02 14:46:53
Message-ID: (view raw, whole thread or download thread mbox)
Lists: pgsql-general

On some corner cases our database servers starts to use from swap space
and that totally makes PostgreSQL (or any other I/O bound application)
to fall upon its knees. In our current structure, responsiveness has the
highest priority and thus it is ok for us to cancel queries at that
instant and re-initiate connections. To achieve this effect, I started
to turn swap space off on some of the servers and turned
vm.oom_kill_allocating_task kernel parameter on. (Periodical postgres
process availability checks decides whether there is a need to fire up a
fresh postgres instance.) So far, this method worked pretty well but I'm
suspicious about data corruptions. (Disks configurations are set to RAID
10.) What are the downsides of such a design scheme?



pgsql-general by date

Next:From: Guillaume LelargeDate: 2008-05-02 14:49:03
Subject: Re: Autovacuum
Previous:From: Scott RibeDate: 2008-05-02 14:29:28
Subject: Re: clustering without locking

Privacy Policy | About PostgreSQL
Copyright © 1996-2018 The PostgreSQL Global Development Group