<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">Leon,<div><span class="Apple-tab-span" style="white-space:pre"> </span>If the backup is to CERNET2, and it's pure v6, what exactly is the plan for v4 connectivity if the primary link goes down? I have two specific concerns:</div><div><br></div><div><span class="Apple-tab-span" style="white-space:pre"> </span>a) That we have reasonably high performance and solid v4 connectivity should the primary ISP link fail</div><div><span class="Apple-tab-span" style="white-space:pre"> </span>b) That the backup path is unfiltered just as we expect of the primary</div><div><br></div><div><span class="Apple-tab-span" style="white-space:pre"> </span>Can you explain a bit more? Perhaps this would be easier once we can see the external connectivity diagram Randy asked for during the tech call.</div><div><br></div><div><span class="Apple-tab-span" style="white-space:pre"> </span>Thanks!</div><div><br></div><div><span class="Apple-tab-span" style="white-space:pre"> </span>- Jim</div><div><br><div><div>On Oct 12, 2010, at 9:06 PM, Xiaoliang Zhao wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite">Hi Jim,<br><br>In our case, the backup link is connecting to CERNET2, which is a pure IPv6 network. in other words, the backup link has very few ipv4 traffic per design. Usual TE tricks may not work well here, i'm afraid.<br>
<br>thanks,<br>Leon<br><br><div class="gmail_quote">On Wed, Oct 13, 2010 at 11:22 AM, Jim Martin <span dir="ltr"><<a href="mailto:jim@daedelus.com">jim@daedelus.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
Leon,<br>
My personal preference is to take full routes from all sources so you have maximum control on our side. The usual approach is to allow simple best path, taking everything from everyone, and letting the usual BGP rules take control. However, you then have the ability to apply LocalPref as a knob to make fine adjustments if you need/want to. This works well if a particular site is having problems over one link and you want to force it out the other. It also allows you to set localpref based upon incoming communities, to say, for example, prefer a R&E path over a commodity.<br>
<br>
- Jim<br>
<div><div></div><div class="h5"><br>
On Oct 12, 2010, at 8:09 PM, Xiaoliang Zhao wrote:<br>
<br>
> Hi,<br>
><br>
> We are thinking to get default route from our ISPs with the intention to simplify the policy. So the basic idea is we get 0/0 from both primary and backup link via bgp, we do local pref to engineer the traffic. Any thoughts on potential issues?<br>
><br>
> thanks,<br>
> Leon<br>
</div></div>> _______________________________________________<br>
> ietf79-tech mailing list<br>
> <a href="mailto:ietf79-tech@daedelus.com">ietf79-tech@daedelus.com</a><br>
> <a href="http://www.daedelus.com/mailman/listinfo/ietf79-tech" target="_blank">http://www.daedelus.com/mailman/listinfo/ietf79-tech</a><br>
<br>
</blockquote></div><br>
</blockquote></div><br></div></body></html>