From: Bodo Eggert <7eggert@gmx.de>
To: Evgeniy Polyakov <zbr@ioremap.net>
Cc: Bodo Eggert <7eggert@gmx.de>,
David Rientjes <rientjes@google.com>,
Alan Cox <alan@lxorguk.ukuu.org.uk>,
linux-kernel@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>,
Linus Torvalds <torvalds@linux-foundation.org>
Subject: Re: Linux killed Kenny, bastard!
Date: Sun, 18 Jan 2009 13:49:58 +0100 (CET) [thread overview]
Message-ID: <alpine.LSU.0.999.0901181337430.6179@be1.lrz> (raw)
In-Reply-To: <20090117154147.GA12863@ioremap.net>
On Sat, 17 Jan 2009, Evgeniy Polyakov wrote:
> On Sat, Jan 17, 2009 at 04:21:50PM +0100, Bodo Eggert (7eggert@gmx.de) wrote:
> > > The only sane way to use oom_adj is
> > > to disable oom killer for the task or make its score very small or very
> > > big, there is really no way to make a finegrained tuning, since score
> > > changes and userspace does not know the algorithm.
> >
> > It does not need to know, it just has to tune until it's about level with
> > the other normal processes if it's to be a normal process, and add or
> > substract one or two to oom_adj if it's avery (un)important process.
>
> Did you try such tuning yourself?
No, I just designed the oom_adj mechanism and left the implementation to
experienced people.
> I submitted documentation update on
> how score is calculated, there is really no way admin can do that in
> some script or manually, and relying on oom_score content is not very
> precise since it jumps all over the range depending on the read time.
The value is expeted to vary with the read time, since memory usage
does vary, too. If your sshd goes berserk and starts eating memory,
it may be smart to kill it. If it doesn't, the score should remain
about the same.
> So, forget that you can tune oom_adj system-wide, it is only possible to
> effectively enable or disable it.
It is, as long as the score varies with the badness of a process.
> If there are two identical processes,
> it is possible in theory to tune them against each other, but practice
> and theory are the same only in theory, in practice one of the processes
> will suddenly start doing different things and all your calculus
> immediately become wrong.
> OOM scores can not be reliably tuned system-wide.
As I said, if a program starts misbehaving, it's mostly not sane to keep
it alive. Besides that, the oom_adj allows you to special case some
processes to be immortal or to be kenny, but you should be prepared for
trouble then.
--
A Purple Heart just goes to prove that were you smart enough to think of a
plan, stupid enough to try it, and lucky enough to survive.
next prev parent reply other threads:[~2009-01-18 12:50 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <bTxPW-1lH-13@gated-at.bofh.it>
[not found] ` <bTE53-5LJ-13@gated-at.bofh.it>
[not found] ` <bTEeI-5Y0-15@gated-at.bofh.it>
[not found] ` <bTGTe-1K5-3@gated-at.bofh.it>
[not found] ` <bTNi6-3kb-9@gated-at.bofh.it>
[not found] ` <bTOea-4QT-1@gated-at.bofh.it>
[not found] ` <bTQ6f-7Qs-1@gated-at.bofh.it>
[not found] ` <bTQpv-5o-3@gated-at.bofh.it>
[not found] ` <bTQzd-iK-5@gated-at.bofh.it>
[not found] ` <bTXhp-2IA-31@gated-at.bofh.it>
[not found] ` <bTZjb-62D-25@gated-at.bofh.it>
2009-01-17 15:21 ` Bodo Eggert
2009-01-17 15:41 ` Evgeniy Polyakov
2009-01-18 12:49 ` Bodo Eggert [this message]
2009-01-18 13:17 ` Evgeniy Polyakov
2009-01-13 10:58 Tomasz Chmielewski
2009-01-13 12:20 ` Evgeniy Polyakov
-- strict thread matches above, loose matches on Subject: below --
2009-01-12 15:33 Evgeniy Polyakov
2009-01-12 15:44 ` Dave Jones
2009-01-12 15:48 ` Evgeniy Polyakov
2009-01-12 15:51 ` Alan Cox
2009-01-12 15:52 ` Evgeniy Polyakov
2009-01-12 21:29 ` Chris Snook
2009-01-12 21:42 ` Evgeniy Polyakov
2009-01-12 15:49 ` Alan Cox
2009-01-12 15:50 ` Evgeniy Polyakov
2009-01-12 15:52 ` Alan Cox
2009-01-12 15:56 ` Evgeniy Polyakov
2009-01-12 16:19 ` Alan Cox
2009-01-12 16:29 ` Evgeniy Polyakov
2009-01-12 23:00 ` Bill Davidsen
2009-01-12 23:17 ` Evgeniy Polyakov
2009-01-13 1:53 ` David Rientjes
2009-01-13 8:52 ` Evgeniy Polyakov
2009-01-13 9:54 ` David Rientjes
2009-01-13 11:54 ` Evgeniy Polyakov
2009-01-13 12:15 ` Alan Cox
2009-01-13 12:29 ` Evgeniy Polyakov
2009-01-13 13:19 ` Theodore Tso
2009-01-13 13:35 ` Evgeniy Polyakov
2009-01-14 0:24 ` Bill Davidsen
2009-01-14 0:35 ` Evgeniy Polyakov
2009-01-13 13:47 ` Alan Cox
2009-01-13 19:36 ` David Rientjes
2009-01-13 21:46 ` Evgeniy Polyakov
2009-01-13 22:49 ` Theodore Tso
2009-01-13 23:02 ` Evgeniy Polyakov
2009-01-14 1:11 ` Theodore Tso
2009-01-14 1:20 ` Evgeniy Polyakov
2009-01-14 4:06 ` Theodore Tso
2009-01-13 23:10 ` David Rientjes
2009-01-13 23:35 ` Evgeniy Polyakov
2009-01-13 23:43 ` David Rientjes
2009-01-13 23:55 ` Evgeniy Polyakov
2009-01-14 0:32 ` David Rientjes
2009-01-14 0:53 ` Evgeniy Polyakov
2009-01-14 4:23 ` Valdis.Kletnieks
2009-01-14 9:07 ` Evgeniy Polyakov
2009-01-13 19:15 ` David Rientjes
2009-01-13 22:00 ` Evgeniy Polyakov
2009-01-13 23:26 ` Valdis.Kletnieks
2009-01-13 23:36 ` Evgeniy Polyakov
2009-01-13 13:41 ` Jan-Frode Myklebust
2009-01-13 13:59 ` Alan Cox
2009-01-12 16:22 ` Dave Jones
2009-01-12 16:28 ` Evgeniy Polyakov
2009-01-13 16:35 ` KOSAKI Motohiro
2009-01-13 22:04 ` Evgeniy Polyakov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=alpine.LSU.0.999.0901181337430.6179@be1.lrz \
--to=7eggert@gmx.de \
--cc=akpm@linux-foundation.org \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=linux-kernel@vger.kernel.org \
--cc=rientjes@google.com \
--cc=torvalds@linux-foundation.org \
--cc=zbr@ioremap.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®