What steps would I need to take in order to obtain 75,000 concurrent TCP
sessions on a FreeBSD 5.2 system running on the following hardware:
dual xenon 3ghz 1mb cache processors
2 gigs of memory
two dual port fibre gigabit nic's
1 onboard copper 10/100 nic
I read a post that was sent to freebsd-
Guy Helmer wrote:
Emre Bastuz wrote on Thursday, February 12, 2004 3:43 PM
Hi,
for sniffing purposes I have a FreeBSD 4.9-STABLE box running on highend,
state-of-the-art hardware (Xeon something) with all bells and whistles.
The NIC´s an onboard copper em0 with gig-e capabilities.
Nevertheles
On Thu, Feb 12, 2004 at 02:27:21PM -0800, Brooks Davis wrote:
> On Thu, Feb 12, 2004 at 02:40:42PM -0500, Brian Reichert wrote:
> > If I catch a kernel doing otherwise, can I say 'Aha! That's a bug
> > based on documented standards' ?
>
> Since Solaris switching to doing round-robin in something l
You can start by doing "netstat -m" and take it from there
~~
Mustafa N. Deeb
Technical Director
Palnet Communications Ltd.
Tel: +970-2-2403434
Fax: +970-2-2403430
www.palsms.com
www.paltime.net
www.palnet.com
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL
On Thu, Feb 12, 2004 at 02:40:42PM -0500, Brian Reichert wrote:
> On Thu, Feb 12, 2004 at 11:35:06AM -0800, Julian Elischer wrote:
> > > My expectation was that the primary IP address would be used.
> >
> >
> > The primary IP address on the interface referred to in the routing table
> > entry tha
Emre Bastuz wrote on Thursday, February 12, 2004 3:43 PM
> Hi,
>
> for sniffing purposes I have a FreeBSD 4.9-STABLE box running on highend,
> state-of-the-art hardware (Xeon something) with all bells and whistles.
>
> The NIC´s an onboard copper em0 with gig-e capabilities.
>
> Nevertheless I am g
Hi,
for sniffing purposes I have a FreeBSD 4.9-STABLE box running on highend,
state-of-the-art hardware (Xeon something) with all bells and whistles.
The NIC´s an onboard copper em0 with gig-e capabilities.
Nevertheless I am getting massive packet drops (40%-60%) when I start sniffing a
gigabit
On Thu, Feb 12, 2004 at 11:35:06AM -0800, Julian Elischer wrote:
> > My expectation was that the primary IP address would be used.
>
>
> The primary IP address on the interface referred to in the routing table
> entry that is chosen for the first packet..
> (last time I looked)
Such was my expec
--- Brian Reichert <[EMAIL PROTECTED]> wrote:
> I've had an expectation violated recently, and
> wanted to research
> whether my expectation was grounded in reality:
>
> Given an interface with a primary IP, and one or
> more aliased IP
> addresses:
>
> When a process opens a connection to
On Thu, 12 Feb 2004, Brian Reichert wrote:
> I've had an expectation violated recently, and wanted to research
> whether my expectation was grounded in reality:
>
> Given an interface with a primary IP, and one or more aliased IP
> addresses:
>
> When a process opens a connection to anot
I've had an expectation violated recently, and wanted to research
whether my expectation was grounded in reality:
Given an interface with a primary IP, and one or more aliased IP
addresses:
When a process opens a connection to another host, which IP address
will be chosen as the source ad
Christian,
I recall reading some problems like this on this list recently. I don't
remember if there was
a solution for NATd, but running IPNAT compiled into the kernel has been
highly efficient
for my small office. I have no problems with transfers up to 3MB/sec.
Maybe higher. I
have also foun
Hi,
I setup a computer to act as a natd for our office. Everything works fine
but I'm trying to tweak it a little bit to get extra speed.
When I download from box itself I easily get 6 or 7 mbytes/sec. but
when I do it behind the nat (office pc). I only get ~ 500k/sec.
Is there a way to tweak t
i want to make my own node with my own specifications.
how can i do that and load it and pass data through
it.
reply as soon as possible...
cheers
manish
Yahoo! India Education Special: Study in the UK now.
Go to http://in.
14 matches
Mail list logo
