Date: Tue, 29 Jan 2002 11:42:17 +0100 (CET) From: Alexander Leidinger <Alexander@Leidinger.net> To: forrestc@imach.com Cc: absinthe@pobox.com, troy@psknet.com, freebsd-isp@FreeBSD.ORG Subject: Re: Tuning a system... Message-ID: <200201291042.g0TAgIs03880@Magelan.Leidinger.net> In-Reply-To: <Pine.BSF.4.21.0201290019360.16777-100000@workhorse.iMach.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 29 Jan, Forrest W. Christian wrote:
>> I'm of the school of thought that says you're likely to spend a lot more time
>> making kernel adjustments, compiling and rebooting than you ever would trying
>> to reason with a dynamic kernel or deal with memory shortages, especially
>> with as cheap as RAM is now.
>
> I'll add my $0.02:
>
> About the only thing which I have run across which I regularly run into
> problems with are TCP/IP buffer parameters such as mbufs. If I tune my
> kernel so I can get insane flows across the internet (DS-3/OC-3 level
> links), I will quickly run out of buffers if I get more than a couple of
> flows going.
>
> I would *love* to see these dynamically sized - the buffer space, not
> necessarily the tcp/ip tuning parameters.
-current (and perhaps 4.5) do something like this already at boot time
depending on the amount of memory (you have to set maxusers to 0 to get
this behavior). It's not dynamic at run time, it's only dynamic at boot
time. And if it doesn't allocate enough for you, you can override it
with your own settings.
Bye,
Alexander.
--
Loose bits sink chips.
http://www.Leidinger.net Alexander @ Leidinger.net
GPG fingerprint = C518 BC70 E67F 143F BE91 3365 79E2 9C60 B006 3FE7
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-isp" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200201291042.g0TAgIs03880>