Message-ID: <46811442.5070308@trash.net>
Date: Tue, 26 Jun 2007 15:27:30 +0200
From: Patrick McHardy <kaber@trash.net>
User-Agent: Mozilla-Thunderbird 2.0.0.0 (X11/20070601)
MIME-Version: 1.0
To: Vasily Averin <vvs@sw.ru>
CC: Eric Dumazet <dada1@cosmosbay.com>,
       "David S. Miller" <davem@davemloft.net>,
       netfilter-devel@lists.netfilter.org, rusty@rustcorp.com.au,
       Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
       devel@openvz.org
Subject: Re: [NETFILTER] early_drop() imrovement (v3)
References: <4615FE1D.80206@sw.ru> <20070406102433.d3a670a5.dada1@cosmosbay.com> <4616203A.80203@sw.ru> <4616626C.9020100@trash.net> <4617845F.7080203@sw.ru> <461789CF.8000106@cosmosbay.com> <46187770.1070106@sw.ru> <46417137.5080501@sw.ru> <467FC8D2.5070102@trash.net> <46811292.1010501@sw.ru>
In-Reply-To: <46811292.1010501@sw.ru>
Content-Type: text/plain; charset=ISO-8859-15; format=flowed
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1634
Lines: 36

Vasily Averin wrote:
> Patrick McHardy wrote:
>> I don't like the NF_CT_PER_BUCKET constant. First of all, each
>> conntrack is hashed twice, so its really only 1/2 of the average
>> conntracks per bucket. Secondly, its only a default and many
>> people use nf_conntrack_max = nf_conntrack_htable_size / 2, so
>> using this constant for early_drop seems wrong.
>>     
>> Perhaps make it 2 * nf_conntrack_max / nf_conntrack_htable_size
>> or even add a nf_conntrack_eviction_range sysctl.
>>     
>
> IMHO The number of conntracks checked in early_drop() have following restrictions:
> - it should be not too low -- to decrease chances of transmission failures,
> - it should be limited by some reasonable value -- to prevent long check delays.

Agreed.

> Also I believe it makes sense to have it constant (how about NF_CT_EVICTION
> name?) -- to have the same behaviour on various nodes. However I doubt strongly
> that anybody will want to change this value. Do you think it is really required?
>   

I don't know. The current behaviour will on average scan 16 entries.
For people manually tuning their hash to saner settings it will scan
a single entry. So we have a quite wide range of values already.
The single entry with sane hash settings is too little IMO, maybe use
some middle-ground, make it 8 by default as you did and rename the
constant. NF_CT_EVICTION_RANGE sounds fine.


-
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/