On Thu, Jul 20, 2006 at 06:31:37PM +0200, Alan Barrett wrote:
> On Thu, 20 Jul 2006, Christos Zoulas wrote:
> > I don't get it. Isn't the old code and the new code equivalent?
> 
> There's a bug in route(8) such that "route delete default ${foo}"
> deletes the default route even if the gateway was not ${foo}.  The
> script under discussion seems to be an attempt to work around the bug.

I am pretty sure the bug is in the kernel, which doesn't match routes by
both destination and gateway, but by destination, only.  After a cursory
look at rtrequest1(), I feel more certain that is where the bug is.

Dave

-- 
David Young             OJC Technologies
dyoung@ojctech.com      Urbana, IL * (217) 278-3933