netfilter-devel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Krzysztof Olędzki" <ole@ans.pl>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Anand Raj Manickam <anandrm@gmail.com>,
	netdev@vger.kernel.org, netfilter-devel@vger.kernel.org,
	shemminger@vyatta.com
Subject: Re: Kernel Panic on OOM with 10 HTB rules
Date: Wed, 08 Sep 2010 17:02:29 +0200	[thread overview]
Message-ID: <4C87A585.5090109@ans.pl> (raw)
In-Reply-To: <1283957137.2748.71.camel@edumazet-laptop>

On 2010-09-08 16:45, Eric Dumazet wrote:
<CUT>
> Thanks
>
> You are running a 32bit kernel, so kernel can _not_ use more than 800
> Mbytes of memory (Lowmem) for his own needs (including all network
> buffers)
>
> If you cannot switch to 64bit kernel, then you are forced to use lower
> queue lengths (I see your imq devices use insane 11000 txqueuelen)
>
> Each frame use 4K, maybe 16K, it depends on MTU.
>
> even if we dont take into account other needs :
> 11000 * 16K = 170 Mbytes per imqX
> 1000 * 4K = 4Mbytes per ethX
>
> 170M * 8 ->  memory overflow
>
> Also /proc/sys/vm/min_free_kbytes  limit is probably too small.
>
> Really, using a 32bit kernel on this machine is not wise.
>
> Even reducing physical ram to 4GB would be better, you would get more
> Lowmem memory ;)
>
> Since vmalloc is not used very much on your machine you can reduce
> vmalloc window from 128M to 32M to save 96MB of low mem
> (boot parameter : vmalloc=32M )

It is also possible to build a kernel using 2G/2G MEM SPLIT or even 
1G/3G if there is small userspace activity. But indeed, x86-64 is the 
proper solution to choose.

Best regards,

			Krzysztof Olędzki

  reply	other threads:[~2010-09-08 15:02 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-09-08 10:52 Kernel Panic on OOM with 10 HTB rules Anand Raj Manickam
2010-09-08 12:42 ` Eric Dumazet
2010-09-08 14:09   ` Anand Raj Manickam
2010-09-08 14:45     ` Eric Dumazet
2010-09-08 15:02       ` Krzysztof Olędzki [this message]
2010-09-08 17:25       ` Jarek Poplawski
2010-09-16  9:35   ` Patrick McHardy
2010-09-16 12:20     ` Anand Raj Manickam

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=4C87A585.5090109@ans.pl \
    --to=ole@ans.pl \
    --cc=anandrm@gmail.com \
    --cc=eric.dumazet@gmail.com \
    --cc=netdev@vger.kernel.org \
    --cc=netfilter-devel@vger.kernel.org \
    --cc=shemminger@vyatta.com \
    /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;
as well as URLs for NNTP newsgroup(s).