A bit of a route socket bug...

[Andrew R. Reiter]

Slightly off topic, but have you had a chance to look at that lock 
recursion bug in the routing code that I had reported awhile ago?  
Admittedly, I never send-pr'd it.  However, I should have the original 
information around... Perhaps I will test.

On Wed, 6 Jul 2005 gnn at freebsd.org wrote:

:Howdy,
:
:As was reported in
:http://www.freebsd.org/cgi/query-pr.cgi?pr=%0D%0A82974
:
:if you do 
:
:route change 10.3.2.1 127.0.0.1
:
:on a system with a default route but no route on the 10 network at
:all, you wind up setting the default route to 127.0.0.1 which is
:rarely what you want.  This is due to the fact that in rtsock.c around
:line 431 (in CURRENT) we do a rnh_lookup but don't make sure that the
:route we looked up was the one we sought.  The radix code correctly
:finds the most useful route to 10.3.2.1, i.e the default route, and
:the dutifully hoses it.
:
:The question is, "What's the right fix here?"  Right now I'm too tired
:to figure it out, so I figure in the 8 hours I'm asleep someone else
:might come up with a patch :-)
:
:Later,
:George
:
:_______________________________________________
:freebsd-net at freebsd.org mailing list
:http://lists.freebsd.org/mailman/listinfo/freebsd-net
:To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"
:
:

--
Andrew R. Reiter
arr at watson.org