Date: Thu, 30 Apr 1998 14:54:59 -0400 From: sbabkin@dcn.att.com To: klmac@baldcom.net Cc: hackers@FreeBSD.ORG Subject: RE: Ethernet address takeover implementation proposal Message-ID: <C50B6FBA632FD111AF0F0000C0AD71EEACDEF4@dcn71.dcn.att.com>
next in thread | raw e-mail | index | archive | help
> ---------- > From: Ken McKittrick[SMTP:klmac@baldcom.net] > > The link-level protocol would go a ways towards the High Availability > project. I more intested in that area. > This is a short description of my implementation (as I remember it, I don't have it handy now to look up): The protocol implementation is send-only. BPF can be used for receiving. I think it's a lot simpler and more elegant way than inventing the second BPF-like interface. BPF already has enough power to filter out any subset of packets received through an interface. The protocol itself does not make any assumptions about the packet contents. It is just transferring the contents to the link level. Of course, the implementation has some link-level-specific parts to extract the link-level address from a packet and pass it to the lower level. Now only the Ethernet support if present. So, essentially it's almost a NOP. :-) Any suggestions/objections are welcome ! :-) Especially from the core team members :-) Besides sending out a broadcast ARP response, other possible usages of this thing include, for example, a slow and inefficient user-level Ethernet bridge with arbitrary firewall. -Serge To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?C50B6FBA632FD111AF0F0000C0AD71EEACDEF4>