LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Toon van der Pas <toon@hout.vanvergehaald.nl>
To: Willy Tarreau <w@1wt.eu>
Cc: Kumar Gala <galak@kernel.crashing.org>,
	Linux Kernel list <linux-kernel@vger.kernel.org>
Subject: Re: tuning/tweaking VM settings for low memory (preventing OOM)
Date: Sat, 13 Jan 2007 14:16:01 +0100	[thread overview]
Message-ID: <20070113131601.GA12901@shuttle.vanvergehaald.nl> (raw)
In-Reply-To: <20070113072217.GW24090@1wt.eu>

On Sat, Jan 13, 2007 at 08:22:18AM +0100, Willy Tarreau wrote:
> > 
> > Which makes me think that we aren't writing back fast enough.  If I  
> > mount the drive "sync" the issue clearly goes away.
> > 
> > It appears from an strace we are doing ftruncate64(5, 178257920) when  
> > we OOM.
> > 
> > Any ideas on VM parameters to tweak so we throttle this from occurring?
> 
> Take a look at /proc/sys/vm/bdflush. There are several useful parameters
> there (doc is in linux-xxx/Documentation). For instance, the first column
> is the percentage of memory used by writes before starting to write on
> disk. When using tcpdump intensively, I lower this one to about 1%.
> 
> Willy

Hi Willy,

I know you're doing a great job on keeping the 2.4 kernel in shape,
but do you also have a good advice for people with more recent
kernels?  (hint: the file /proc/sys/vm/bdflush is missing)

Regards,
Toon.

$ uname -a
Linux shuttle 2.6.18-gentoo-r6 #1 Sat Jan 13 12:00:15 CET 2007 x86_64 AMD Athlon(tm) 64 Processor 3200+ AuthenticAMD GNU/Linux
$ ls -l /proc/sys/vm/
totaal 0
-rw-r--r-- 1 root root 0 jan 13 14:09 block_dump
-rw-r--r-- 1 root root 0 jan 13 14:09 dirty_background_ratio
-rw-r--r-- 1 root root 0 jan 13 14:09 dirty_expire_centisecs
-rw-r--r-- 1 root root 0 jan 13 14:09 dirty_ratio
-rw-r--r-- 1 root root 0 jan 13 14:09 dirty_writeback_centisecs
-rw-r--r-- 1 root root 0 jan 13 14:09 drop_caches
-rw-r--r-- 1 root root 0 jan 13 14:09 hugetlb_shm_group
-rw-r--r-- 1 root root 0 jan 13 14:09 laptop_mode
-rw-r--r-- 1 root root 0 jan 13 14:09 legacy_va_layout
-rw-r--r-- 1 root root 0 jan 13 14:09 lowmem_reserve_ratio
-rw-r--r-- 1 root root 0 jan 13 14:09 max_map_count
-rw-r--r-- 1 root root 0 jan 13 14:09 min_free_kbytes
-rw-r--r-- 1 root root 0 jan 13 14:09 nr_hugepages
-r--r--r-- 1 root root 0 jan 13 14:09 nr_pdflush_threads
-rw-r--r-- 1 root root 0 jan 13 14:09 overcommit_memory
-rw-r--r-- 1 root root 0 jan 13 14:09 overcommit_ratio
-rw-r--r-- 1 root root 0 jan 13 14:09 page-cluster
-rw-r--r-- 1 root root 0 jan 13 14:09 panic_on_oom
-rw-r--r-- 1 root root 0 jan 13 14:09 percpu_pagelist_fraction
-rw-r--r-- 1 root root 0 jan 13 14:09 swappiness
-rw-r--r-- 1 root root 0 jan 13 14:09 swap_token_timeout
-rw-r--r-- 1 root root 0 jan 13 14:09 vfs_cache_pressure

  reply	other threads:[~2007-01-13 13:16 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-01-12 21:58 Kumar Gala
2007-01-13  4:54 ` Nick Piggin
2007-01-15 17:13   ` Kumar Gala
2007-01-13  7:22 ` Willy Tarreau
2007-01-13 13:16   ` Toon van der Pas [this message]
2007-01-13 14:30     ` Willy Tarreau
2007-01-13 23:58       ` Toon van der Pas

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=20070113131601.GA12901@shuttle.vanvergehaald.nl \
    --to=toon@hout.vanvergehaald.nl \
    --cc=galak@kernel.crashing.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=w@1wt.eu \
    --subject='Re: tuning/tweaking VM settings for low memory (preventing OOM)' \
    /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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).