From owner-freebsd-bugs Sat Feb 2 13:10:19 2002 Delivered-To: freebsd-bugs@hub.freebsd.org Received: from freefall.freebsd.org (freefall.FreeBSD.org [216.136.204.21]) by hub.freebsd.org (Postfix) with ESMTP id 7A04237B417 for ; Sat, 2 Feb 2002 13:10:02 -0800 (PST) Received: (from gnats@localhost) by freefall.freebsd.org (8.11.6/8.11.6) id g12LA2944227; Sat, 2 Feb 2002 13:10:02 -0800 (PST) (envelope-from gnats) Date: Sat, 2 Feb 2002 13:10:02 -0800 (PST) Message-Id: <200202022110.g12LA2944227@freefall.freebsd.org> To: freebsd-bugs@FreeBSD.org Cc: From: Archie Cobbs Subject: Re: kern/31586: netgraph bridges cause connectivity probs from bridge Reply-To: Archie Cobbs Sender: owner-freebsd-bugs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org The following reply was made to PR kern/31586; it has been noted by GNATS. From: Archie Cobbs To: freebsd-gnats-submit@FreeBSD.org, michaelb@thus.net Cc: Subject: Re: kern/31586: netgraph bridges cause connectivity probs from bridge Date: Sat, 02 Feb 2002 12:47:36 -0800 I think the problem is the delayed checksum support added to FreeBSD. The delayed checksum code makes the (reasonable) assumption in ip_output() that a packet routed to a specific interface will actually be transmitted by the hardware associated with that interface. This is not necessarily true when netgraph is being used. The solution would be to modify ng_ether() to compute on-demand any delayed checksums for packets leaving the 'lower' hook when that hook is attached to another node. I'll try to work on this soon. -Archie __________________________________________________________________________ Archie Cobbs * Packet Design * http://www.packetdesign.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-bugs" in the body of the message