<html>
<head>
<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<div class="moz-cite-prefix">On 01/17/2015 02:24 PM, Henning Rogge
wrote:<br>
</div>
<blockquote
cite="mid:CAGnRvup+=QU6y1+pcYMDDvpBWaJ_iL4obmz04kODxOAKBxfNTg@mail.gmail.com"
type="cite">
<pre wrap="">On Sun, Jan 11, 2015 at 7:52 PM, Nemesis <a moz-do-not-send="true" class="moz-txt-link-rfc2396E" href="mailto:nemesis@ninux.org"><nemesis@ninux.org></a> wrote:
</pre>
<blockquote type="cite" style="color: #000000;">
<pre wrap=""><span class="moz-txt-citetags">> </span>Hi Thijs,
<span class="moz-txt-citetags">></span>
<span class="moz-txt-citetags">> </span>On 01/11/2015 06:57 PM, Thijs van Veen wrote:
<span class="moz-txt-citetags">> </span>...
</pre>
<blockquote type="cite" style="color: #000000;">
<pre wrap=""><span class="moz-txt-citetags">>> </span>1) In the Route object, you suggest a Device field to reference the
<span class="moz-txt-citetags">>> </span>actual interface you're targeting.
<span class="moz-txt-citetags">>> </span>I suggest changing the name Device to Destination Interface or something
<span class="moz-txt-citetags">>> </span>similar to differentiate in terminology between devices (nodes) and
<span class="moz-txt-citetags">>> </span>their interfaces.
</pre>
</blockquote>
</blockquote>
<pre wrap="">Incoming/Outgoing interface maybe?
Source/Destination sounds more like end-to-end information.
</pre>
<blockquote type="cite" style="color: #000000;">
<blockquote type="cite" style="color: #000000;">
<pre wrap=""><span class="moz-txt-citetags">>> </span>2.) You suggest a Source field for source-specific routing.
<span class="moz-txt-citetags">>> </span>I suggest adding a Source Interface field, this way (it seems to me) you
<span class="moz-txt-citetags">>> </span>can more easily handle multi-interface devices.
</pre>
</blockquote>
</blockquote>
<pre wrap="">We could add it as an optional field, not sure if the Linux kernel
supports it at the moment.
But it could be simulated by multiple routing tables...
Yes, I think we should do this.</pre>
</blockquote>
<br>
Henning, thank you for your help in this, especially in ther routing
part.<br>
<br>
<font face="Helvetica, Arial, sans-serif">I must find the time to
collect all the feedback we got <font face="Helvetica, Arial,
sans-serif">recently from different sources<font
face="Helvetica, Arial, sans-serif"> <font face="Helvetica,
Arial, sans-serif">so we can<font face="Helvetica, Arial,
sans-serif"> decide how to proceed.</font></font></font><br>
<br>
<font face="Helvetica, Arial, sans-serif">Peraphs I could write
it <font face="Helvetica, Arial, sans-serif">on a etherpad<font
face="Helvetica, Arial, sans-serif"> or something similar.<br>
<br>
<font face="Helvetica, Arial, sans-serif">The step af<font
face="Helvetica, Arial, sans-serif">ter this one will
be <font face="Helvetica, Arial, sans-serif">to start
implementing it<font face="Helvetica, Arial,
sans-serif"> and <font face="Helvetica, Arial,
sans-serif">get feedback from real world sc<font
face="Helvetica, Arial, sans-serif">enarios.</font></font></font></font></font></font><br>
<br>
Feder<font face="Helvetica, Arial, sans-serif">ico</font><br>
</font></font></font><font face="Helvetica, Arial,
sans-serif"></font></font></font>
</body>
</html>