<br><br><div class="gmail_quote">On Tue, Oct 4, 2011 at 1:46 PM, Benny Baumann <span dir="ltr"><<a href="mailto:BenBE1987@gmx.net">BenBE1987@gmx.net</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
Hi Markus,<br>
<br>
Am 04.10.2011 11:23, schrieb Markus Kittenberger:<br>
<div class="im">> On Tue, Oct 4, 2011 at 2:00 AM, Benny Baumann <<a href="mailto:BenBE1987@gmx.net">BenBE1987@gmx.net</a><br>
</div><div class="im">> <mailto:<a href="mailto:BenBE1987@gmx.net">BenBE1987@gmx.net</a>>> wrote:<br>
><br>
> Hi guys,<br>
> (well, several of our routers died regularly from the routing tables<br>
> being too large).<br>
><br>
> hmm why are the routing tables getting too large?<br>
</div>When doing FIBMetric "correct" the RT not only holds one route per<br></blockquote><div>use flat! *G </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
destination, but one route per distance and host. So sometimes we get up<br>
to 2 or 3 routes per host.<br></blockquote><div>(which is (or at least sounds like) a bug,..) </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div class="im">> olsrd would anyways just create as many routes as there are targets<br>
> (hosts)<br>
> (regardless of the topology)<br>
</div>Yes, but using the plugin we can somehow channel these routes somewhat<br>
and make some of them appear as "INFINITE" so OLSR knows the host is<br>
there, but doesn't try to route directly to it. And that's what wwe<br>
wanted to get.<br></blockquote><div>knowing a host/link is there, also needs memory (just less than having it in topology and creating a route for it) </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div class="im">> but the with such a huge layer2 broadcast domain, olsrds network<br>
> topology gets bigger (and therefore the olsrd memeory consumption)<br>
> furthermore the number of olsrd messages might get very unfunny)<br>
</div>Problems arose for us at only about 10-15 nodes in the Tinc network. A<br>
live graph can be seen at <a href="http://graph.chemnitz.freifunk.net/ffc.svg" target="_blank">http://graph.chemnitz.freifunk.net/ffc.svg</a> The<br>
plugin doesn't run on all the (pink) nodes yet, but on those which have<br>
it you'll see no gray lines (which are the virtual connections Tinc<br>
introduces). The pink lines are direct Tinc connections. BTW: This<br>
drastically reduces the size of TC messages which are much more common<br></blockquote><div>sure the tc messages are a bigger traffic/bandwidth problem as the hellos.</div><div>(but they can be fragmented into multiple messages)</div>
<div><br></div><div>so as hellos can not fragmented they limit your maximum number of neighbours,..</div><div>but for ipv4 its more than 100 neighbours that fit into one packet,..</div><div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
than the HELLO messages (I sent the stats for our network previously on<br>
the list). For all the Tinc nodes we use HELLO intervals of sometimes up<br>
to 5 minutes thus there's not much concern regarding HELLO messages;<br></blockquote><div>ok, fine. but as said above its not only about bandwidth,.. </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
more regarding TC. Currently we have 4.5KB/s TC down and 0.5KB/s TC up;<br>
which makes using the network on Edge links feasable (not quite nice,<br>
but feasable).<br></blockquote><div>this sound like a very small network,.. (but i don not know your tc timings on all the nodes)</div><div>(but how does tinc broadcast the messages, sending duplicated packets to all other nodes in the tinc cloud?)</div>
<div><br></div><div>would this create 450KB/sec for a 100 nodes tinc cloud??</div><div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div class="im">But as far as I understand it OLSR doesn't include INFINITE/non-SYMM</div>
links in the HELLO-message.</blockquote><div>this is true for Tcs but not for Hellos. </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"> And as we are dropping default quality down<br>
to INFINITE those get removed.</blockquote><div> </div><div>Markus</div></div>