Re: configurability of OOM killer - Mailing list pgsql-hackers

From Tom Lane
Subject Re: configurability of OOM killer
Date
Msg-id 18162.1201910904@sss.pgh.pa.us
Whole thread Raw
In response to configurability of OOM killer  (Alvaro Herrera <alvherre@commandprompt.com>)
Responses Re: configurability of OOM killer  (Andrew Dunstan <andrew@dunslane.net>)
Re: configurability of OOM killer  (Simon Riggs <simon@2ndquadrant.com>)
Re: configurability of OOM killer  (Ron Mayer <rm_pg@cheapcomplexdevices.com>)
Re: configurability of OOM killer  (Jeff Davis <pgsql@j-davis.com>)
List pgsql-hackers
Alvaro Herrera <alvherre@commandprompt.com> writes:
> This page
> http://linux-mm.org/OOM_Killer

Egad.  Whoever thought *this* was a good idea should be taken out
and shot:
   The independent memory size of any child (except a kernel thread) is added to the score:
       /*        * Processes which fork a lot of child processes are likely        * a good choice. We add the vmsize
ofthe childs if they        * have an own mm. This prevents forking servers to flood the        * machine with an
endlessamount of childs        */
 

In other words, server daemons are preferentially killed, and the parent
will *always* get zapped in place of its child (since the child cannot
have a higher score).  No wonder we have to turn off OOM kill.
        regards, tom lane


pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: and waiting
Next
From: Andrew Dunstan
Date:
Subject: Re: configurability of OOM killer