From owner-freebsd-questions Tue Mar 14 6:46:28 2000 Delivered-To: freebsd-questions@freebsd.org Received: from hotmail.com (oe17.law4.hotmail.com [216.33.148.121]) by hub.freebsd.org (Postfix) with SMTP id 8A94B37B777 for ; Tue, 14 Mar 2000 06:46:25 -0800 (PST) (envelope-from orinoki@hotmail.com) Received: (qmail 6160 invoked by uid 65534); 14 Mar 2000 14:46:24 -0000 Message-ID: <20000314144624.6159.qmail@hotmail.com> X-Originating-IP: [62.0.146.166] From: "Indiana Jones" To: Subject: Problems using Freebsd kernel for IP forwarding Date: Tue, 14 Mar 2000 16:35:20 +0200 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0256_01BF8DD3.49E40780" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2014.211 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2014.211 Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG This is a multi-part message in MIME format. ------=_NextPart_000_0256_01BF8DD3.49E40780 Content-Type: text/plain; charset="windows-1255" Content-Transfer-Encoding: quoted-printable Hi, We use a freebsd machine as a simple router with forwarding turned = on. Can someone explain the following kernel behavior: 1. we do 'route add 10.9.109.16/28 192.168.1.1' 2. then someone forwards traffic to address 10.9.109.21 = through this machine which correctly use this routing entry. 3. we then do 'route add 10.9.109.16/29 192.168.1.2' now, when forwarding traffic again to the same IP address = (10.9.109.21) the forwarding is still being done according to the less specific forwarding rule (1) and not to the = best match (3) rule (according to CIDR policy). If the two rules are added w/o traffic being forwarded in the middle = (i.e. w/o step 2), then the behavior is correct. This seems like a bug!? Thanks Orinoki. ------=_NextPart_000_0256_01BF8DD3.49E40780 Content-Type: text/html; charset="windows-1255" Content-Transfer-Encoding: quoted-printable

Hi,

We use = a freebsd=20 machine as a simple router with forwarding turned on.

Can = someone=20 explain the following kernel behavior:

= =20 1. we do 'route add 10.9.109.16/28 = 192.168.1.1'

2. &n= bsp; then=20 someone forwards traffic to address 10.9.109.21 through

&nbs= p; this=20 machine which correctly use this routing entry.

= =20 3. we then do 'route add 10.9.109.16/29 192.168.1.2'

now, = when=20 forwarding traffic again to the same IP address (10.9.109.21) the = forwarding is=20 still being done

= according to the=20 less specific forwarding rule (1) and not to the best match (3) rule = (according=20 to CIDR policy).

If the = two rules=20 are added w/o traffic being forwarded in the middle (i.e. w/o step 2), = then the=20 behavior is correct.

This = seems like a=20 bug!?

Thanks

= =20 = Orinoki.

------=_NextPart_000_0256_01BF8DD3.49E40780-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message