Brilliant! - thanks, that's exactly the info I needed to know. I'm using
the Measurement Factory distribution of FreeBSD4.3 (from the
downloadable ISO) and I know it will allow the aliases I want as it
worked fine on one box - I'm all ready to roll now!
Cheers
Bryn
-----Original Message-----
From: Alex Rousskov [mailto:rousskov@measurement-factory.com]
Sent: 12 March 2003 17:57
To: Bryn Reeves
Cc: users@web-polygraph.org
Subject: Re: aka on FreeBSD
On Wed, 12 Mar 2003, Bryn Reeves wrote:
> Sorry to pester again but I've a small problem with our polygraph
> setup. I'm using the perl scripts and 'aka' tool to do IP aliasing.
>
> Our setup is a class B net with addresses in 10.6.n.m and a netmask of
> 255.255.0.0 - this is set explicitly in the ifconfig lines of rc.conf
on
> all hosts.
>
> On one host, aliases are being created with the correct netmask of
> 255.255.0.0 however the other machine is using 255.255.255.0
> (class-c) which is wrong!
>
> Where does aka pick up netmask settings from? The primary IP of the
> interface concerned has correct netmask, but the aliases don't. I'm
> pretty sure I've just confused things somewhere along the line (it's
> taken some tinkering to get to this stage and I've probably broken
> something!)
You can specify the subnet explicitly with aka (or PGL). For example,
aka lo0 172.16.1.1-10/16
If there is no explicit subnet, Polygraph tries to guess it based on
the number of ranges in the address specification. For more than one
range (more than one dash), /16 is used. Otherwise, /24 is used.
Please keep in mind that recent FreeBSD releases may insist on /32
subnets used for Polygraph-like aliases.
Alex.
--
| HTTP performance - Web Polygraph benchmark
www.measurement-factory.com | HTTP compliance+ - Co-Advisor test suite
| all of the above - PolyBox appliance
This archive was generated by hypermail 2b29 : Mon Feb 06 2006 - 12:00:26 MST