Date: Tue, 9 Sep 2008 14:21:19 +0200 (CEST) From: Wojciech Puchar <wojtek@wojtek.tensor.gdynia.pl> To: Mario Lobo <mlobo@digiart.art.br> Cc: freebsd-questions@freebsd.org Subject: Re: [OT] tcpdump overhead Message-ID: <20080909141945.B37476@wojtek.tensor.gdynia.pl> In-Reply-To: <30719919.1220960726357.JavaMail.nfsnobody@mailapp02.brturbo.com> References: <30719919.1220960726357.JavaMail.nfsnobody@mailapp02.brturbo.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> make is how much overhead tcpdump will have on them (if any). It seems > to me that this overhead can be disregarded as influential but, if depends how kmuch packets are actually gathered. if you tcpdump everything from loaded gigabit link it won't be small. you will have to use at least 2-disk stripping to be able to write data ;) i don't know if it's "official" docs, just generate some traffic, observe how much CPU is idle, then add tcpdump to the mix and do the math. simple
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20080909141945.B37476>