[sdnog] using RIPE atlas to check which DNS root your resolver prefers.

[Tarig Yassin]

I understand your concerns,  but our customers should have access to SIXP even If they are from outside.


--
Tarig YassinHiba Eltigani <higba6 at gmail.com> wrote:
   Ok, I am not sure this will work out, it will need careful configuration
because you don't want to be in a position where Ethionet traffic use the
IXP as a way to reach other opetators which can cause bandwidth issue. I
guess this is where L2 IXP comes handy.
   I can't recall the exact case but one of the big IXPs had issue after
their local prefix leaked to the Internet.
BR

Hiba
On Dec 27, 2014 11:22 PM, "tariq198487" <tariq198487 at hotmail.com> wrote:

> Accordingly, three steps are required for Ethiopian to use our e_root:
> 1- Mohamed Osama ( SIXP admin)  should update his lists with ethiopian
> prefixes.
> 2- commuinty tags should be removed from e-root prefixes.
> 3- operators advertize e-root prefixes to Ethiopian.
>
> Thanks
>
>
>
>
> --
> Tarig Yassin
>
> Hiba Eltigani <higba6 at gmail.com> wrote:
>
> Thanks for sharing. The Ethionet seems really odd but I guess since the
> e.root is hosted in the IXP, most probably the ISP is tagging all the
> routes with no-advertise community to prevent them from going to the
> Internet. I think it is a good practice to keep routes received from IXP
> inside your network but may be root servers can be the exception ;).
> BR
>
> Hiba
> On Dec 27, 2014 5:57 PM, "Nishal Goburdhan" <nishal at controlfreak.co.za>
> wrote:
> >
> >
> > some of you may remember me showing some graphs from the RIPE atlas
> probes, showing DNS performance.  we've distributed several probes in
> sudan, and i thought it would be interesting for you to see what these can
> show you about the networks that the probes are attached to.
> >
> > if you look at:
> https://atlas.ripe.net/contrib/root_anycast.html?msm_id=1 and choose
> "e.root" and "IPv4", you'll see that all the results for the probes in
> sudan, report KRT.  that's good;  it tells us that from the different
> network operators that have probes installed, DNS responses to e.root goes
> "local" to khartoum.  this is what we want!
> >
> > in looking at this map, i noticed some oddities;  for example, if you
> look at the probes in ethiopia, you'll see a response for SDB;  which is a
> location in cape town.  so DNS resolution for the probes in those ethiopian
> networks, goes all the way to cape town.  that's clearly not very optimal.
> > what's likely happening, is that ethiopia telecom, buys IP transit from
> seacom, who also buy transit from a south african ISP, who advertises the
> e.root prefix to seacom.  seacom then route down to south africa to what
> they believe to be the closest e.root.
> >
> > i call this odd, since, at least one sudanese operator is selling IP
> transit to ethio telecom, so i would expect the ethio telecom probe to also
> pickup the e.root from khartoum.  perhaps if the operator is here, and
> listening, they can look into this  ;-)
> >
>
> > you can have a lot more fun (and learn a lot more about how things route
> across the 'net, using the RIPE atlas probes.  you'll see that there aren't
> a lot of probes installed in sudan;  it would be really good to get this
> from as many (network) locations  as possible.  so if you want to help host
> a probe, please ping me (or sara) off-list.
> >
> > --n.
> > _______________________________________________
> > Sdnog mailing list
> > Sdnog at sdnog.sd
> > http://lists.sdnog.sd/mailman/listinfo/sdnog
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sdnog.sd/pipermail/sdnog/attachments/20141228/171d1df4/attachment.html>