From: Adayadil Thomas <adayadil.thomas@gmail.com>
To: Bill Davidsen <davidsen@tmr.com>
Cc: linux-kernel@vger.kernel.org
Subject: Re: LowFree pattern
Date: Wed, 2 Sep 2009 18:20:00 -0400 [thread overview]
Message-ID: <fb7befa20909021520o110b9159led09237e72da469e@mail.gmail.com> (raw)
In-Reply-To: <h7mgl3$v6$1@ger.gmane.org>
> What problem did you see that you traced to LowFree?
I have set this device as a bridge. Higher latency and low throughput seem
to be related to LowFree. There are also packet drops at the interface
when LowFree is too low.
I have another device which has 4G RAM split into 900M LowMem and
the rest as HighMem. In that case also, LowMem goes to a low level, say 8M
and although HighMem has ~2G free, the system gives OOPS (related to
page_alloc failures)
when acting as a bridge (passing packets).
On Wed, Sep 2, 2009 at 3:22 PM, Bill Davidsen<davidsen@tmr.com> wrote:
> Adayadil Thomas wrote:
>>
>> Greetings.
>>
>> I am running a centos linux with 2.6.20 version kernel. The system has
>> 1G of RAM.
>>
>> As time goes by the LowFree becomes really low. Right now it shows
>> 137M .. but it goes as low as 8M or so
>>
>> The command -
>> echo 3 > /proc/sys/vm/drop_caches
>> brings back the LowFree to way up high.
>>
>> The question I have is whether the system by itself release the cache
>> (drop cache) automatically
>> to maintain a reasonable LowFree?
>> Is this configurable?
>>
> You can change the parameters in /rpoc/sys/vm if you wish, but what makes
> you think this is needed? Cache is dropped as memory is needed, and
> drop_cache in general is a good way to slow the system.
>
>> Any information or help is much appreciated.
>>
> What problem did you see that you traced to LowFree?
>
>>
>> cat /proc/meminfo
>> MemTotal: 1034788 kB
>> MemFree: 138240 kB
>> Buffers: 99260 kB
>> Cached: 177776 kB
>> SwapCached: 51740 kB
>> Active: 605172 kB
>> Inactive: 113572 kB
>> HighTotal: 130720 kB
>> HighFree: 252 kB
>> LowTotal: 904068 kB
>> LowFree: 137988 kB
>> SwapTotal: 1048568 kB
>> SwapFree: 976332 kB
>> Dirty: 380 kB
>> Writeback: 0 kB
>> AnonPages: 441348 kB
>> Mapped: 15540 kB
>> Slab: 146088 kB
>> SReclaimable: 104288 kB
>> SUnreclaim: 41800 kB
>> PageTables: 1596 kB
>> NFS_Unstable: 0 kB
>> Bounce: 0 kB
>> CommitLimit: 1565960 kB
>> Committed_AS: 590576 kB
>> VmallocTotal: 114680 kB
>> VmallocUsed: 15052 kB
>> VmallocChunk: 99348 kB
>
>
> --
> Bill Davidsen <davidsen@tmr.com>
> "We have more to fear from the bungling of the incompetent than from
> the machinations of the wicked." - from Slashdot
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
next prev parent reply other threads:[~2009-09-02 22:20 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-09-02 16:32 LowFree pattern Adayadil Thomas
2009-09-02 19:22 ` Bill Davidsen
2009-09-02 22:20 ` Adayadil Thomas [this message]
2009-09-03 13:46 ` Bill Davidsen
2009-09-02 19:23 ` Bill Davidsen
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=fb7befa20909021520o110b9159led09237e72da469e@mail.gmail.com \
--to=adayadil.thomas@gmail.com \
--cc=davidsen@tmr.com \
--cc=linux-kernel@vger.kernel.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.