On Sun, 2009-02-22 at 15:29 -0500, Chuck Lever wrote: > On Feb 22, 2009, at Feb 22, 2009, 2:09 PM, Trond Myklebust wrote: > > I really dislike this idea of adding routing information at the sunrpc > > level. > > Well I think we have had problems in the past where RPC replies on > multi-homed systems go to the wrong interface and are dropped. Isn't > that why we now have the source address field in rpc_create_args? I fail to see the relevance to this case. We're not talking about RPC callbacks here. > I suspect we might have similar problems with NSM, for instance. See Tom Talpey's talk at Connectathon a few years ago. This is why we now have the 'nsm_use_hostnames' sysctl, and why the statd callback code saves FQDNs rather than ip addresses. Trond -- To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
