Re: Snort 2.9.0.3 & Phil Wood's modified libpcap

[Mike Lococo <mikelococo () gmail com>]

On 02/08/2011 10:06 AM, Weir, Jason wrote:
> The change file (goes back to 1994)
> http://www.tcpdump.org/libpcap-changes.txt doesn't mention mmap or
> ring buffer.
>
> Anyone have definitive proof that the latest libpcap versions have
> the good stuff included

That changes file looks to be out of date.  Check the October 27, 2008
1.0.0 changelog below which notes "Support for memory-mapped capture on
Linux":

https://github.com/mcr/libpcap/blob/3c13ac2cc3e06899a8ed1aca3e88b2abebb02c9a/CHANGES

Although I'll echo the warning of other folks that the buffer-size is
hardcoded with daq-0.5 and earlier (although the change in 0.5.1 is news
to me, I look forward to testing it) and too small for a reasonably
sized network.  Packet loss occurs at 100-150mbits (down from 200-300)
when using mmaped capture on libpcap-1.0.0/daq-0.5 compared to
libpcap-0.9.8.  AFPacket is much better with an adequately sized buffer,
and maybe mmap with a reasonable buffer will be good as well.

Cheers,
Mike Lococo

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb
_______________________________________________
Snort-users mailing list
Snort-users () lists sourceforge net
Go to this URL to change user options or unsubscribe:
https://lists.sourceforge.net/lists/listinfo/snort-users
Snort-users list archive:
http://www.geocrawler.com/redir-sf.php3?list=snort-users