Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 10 Mar 2006 22:06:38 +0200 (EET)
From:      Dmitry Pryanishnikov <dmitry@atlantis.dp.ua>
To:        Peter Jeremy <peterjeremy@optushome.com.au>
Cc:        Kostik Belousov <kostikbel@gmail.com>, Michael Proto <mike@jellydonut.org>, freebsd-stable@freebsd.org
Subject:   Re: RELENG_4 on flash disk and swap
Message-ID:  <20060310215306.P83075@atlantis.atlantis.dp.ua>
In-Reply-To: <20060310193248.GC688@turion.vk2pj.dyndns.org>
References:  <20060302181625.I3905@atlantis.atlantis.dp.ua> <76FAD2DB-CD18-42D4-95C8-F016CFB17B00@segpub.com.au> <20060303110936.R86586@atlantis.atlantis.dp.ua> <20060303185157.GB692@turion.vk2pj.dyndns.org> <20060304001224.G356@atlantis.atlantis.dp.ua> <20060304065138.GD692@turion.vk2pj.dyndns.org> <20060310121758.S80837@atlantis.atlantis.dp.ua> <20060310123942.GI37572@deviant.kiev.zoral.com.ua> <20060310153737.X40396@atlantis.atlantis.dp.ua> <20060310193248.GC688@turion.vk2pj.dyndns.org>

next in thread | previous in thread | raw e-mail | index | archive | help

Hello!

On Sat, 11 Mar 2006, Peter Jeremy wrote:
>> But AFAIK the kernel kills NOT the requesting process but the one with the
>> largest RSS. This selection algorithm seems to be the dumbest one, since
>> process with largest RSS almost always is the process which does some real
>> work.
>
> This frees up the greatest amount of memory and so minimises the risk
> of the problem recurring.

  For OS yes, it's great to obtain a lot of free pages. But as I wrote in 
another post, if this process was e.g. bgpd, router becomes unusable
and unreachable. So at this point OS benefit differs from system administrator 
benefit: for me, it would be better if that remote router paniced and rebooted
instead of being up but non-operational.

>> Compare "never satisfy request" and "kill another, totally unrelated,
>> process".
>
> This has been argued about many times in the past - though not with
> any satisfactory solution AFAIK.  Look for 'SIGDANGER' in the archives.

  OK, I see, it's very difficult task to decide which process to kill.
So I as an administrator want to have a tool to turn this process killing
off. Of course I can (and will) patch OS kernel, but I'm sure that I'm not
alone. That's why I've proposed some ideas in this area.

Sincerely, Dmitry
-- 
Atlantis ISP, System Administrator
e-mail:  dmitry@atlantis.dp.ua
nic-hdl: LYNX-RIPE



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060310215306.P83075>