btw nice SGW gets finally tested/debugged/fixed,..<br><br>and ACK, i would not really consider SGW as "very" stable in olsrd,..<br>but ...<br><div><br><div class="gmail_quote">On Tue, Dec 13, 2011 at 10:31 AM, Ferry Huberts <span dir="ltr"><<a href="mailto:mailings@hupie.com">mailings@hupie.com</a>></span> wrote:<br>
how about changing the default to 67 percent for the threshold?<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
we would really like to change that.<br></blockquote><div><br></div><div>... afair the main idea/reasoning for SmartGateways in first place, was to provide a solution for users suffering from changing gateways (which are doing NAT), leading to changing "public" ips.</div>
<div>(i.e the same issue Nat-Threshold was aimed at, but ...)</div><div><br></div><div>or are there new (widely/predominantly used) SGW usecases, i am not aware of?</div><div><br></div><div>imho the "only" behaviour matching the original goal, is to change the gateway only if the old one ist lost (or maybe got extremely bad/costly),.. </div>
<div>i.e. 0 (or a very low value) would be the "correct" default for the SGW-Threshold</div><div><br></div><div>but imho 0.67 is not!<br>as it might even cause switching between gateways every some minutes in some spots in some networks,..</div>
<div><br></div><div>furthermore i really do not get the point of wanting the gateway to switch that early,. #2<br></div><div><br></div><div>as imho configuring the SGW-Threshold too close to 1 would produce more or less the same routing results as without using smart-gateway at all! #1</div>
<div><br></div><div>Markus</div><div><br></div><div>#1 sure its still far better to turn off NAT-Threshold (and its loop potential) and use an quite early switching smartgateway, <br><br>also as it will only switch your own gateway, not disturbing other nodes/users,.<br>
<br></div><div>And btw in mixed network with "old" nodes having NAT-Threshold active, SGW is great for "new" nodes too, as they do not suffer from the wrong/looping default routes a active NAT-Threshold might/will cause on the "old" nodes</div>
<div><br></div><div>#2 of course with multiple tunnels and having the old connections stick to the old tunnel(s) it would be different, but imho this is only doable with kernel/nat integration of these SGWs</div><div><br>
</div></div></div>