Subject: Re: non-standard way to capture network traffic
To: Vlad GALU <vladgalu@gmail.com>
From: Zeljko Vrba <zvrba@globalnet.hr>
List: tech-kern
Date: 09/25/2005 08:14:07
This is an OpenPGP/MIME signed message (RFC 2440 and 3156)
--------------enig82E0D923255FFFA1729AF536
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

Vlad GALU wrote:
> 
>  Since NetBSD uses an unified cache for I/O and VM, your disk writes
> would get delayed a bit, no matter what way you write to your
>
does this happen even if using raw device (/dev/ld0x ?) I was planning
to use large (2 GB) in-memory buffer in a ring fashion, with parallel
writing to disk from another thread.

>
> the trick. As for compression, hmm, depends on what you want to write
> (plaintext output or binary packet headers) and the amount of written
>
I'll be compressing just binary packet headers. it's faster both for
writing and subsequent processing.

>
> data. No one can give you exact figures. It's up to you to test and
> decide.
>
I've searched google for measurements, but came up with nothing. I can't
believe someone didn't already perform the benchmarks. Maybe I just
can't use google properly :/

thanks for the advice :)

--------------enig82E0D923255FFFA1729AF536
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFDNkA3UIHQih3H6ZQRAtvRAKDHLwyGc3lbr/UoiFWCA6TJX/3HDACfU1Zs
8P7HdyJSF3yHRC0FmBeO3Ug=
=hK2K
-----END PGP SIGNATURE-----

--------------enig82E0D923255FFFA1729AF536--