[59896] in SIPB IPv6
Re: occaid pop down?
daemon@ATHENA.MIT.EDU (Erik Nygren)
Fri Oct 23 14:44:43 2015
In-Reply-To: <smuy4evizej.fsf@linuxpal.mit.edu>
Date: Fri, 23 Oct 2015 14:44:41 -0400
From: Erik Nygren <erik@nygren.org>
To: Greg Troxel <gdt@lexort.com>
Cc: =?UTF-8?Q?Alejandro_R=2E_Sede=C3=B1o?= <asedeno@mit.edu>, sipbv6@mit.edu
--001a11330cba22b02d0522ca0194
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Looks like this was fixed around 8pm last night or a little before.
Side-note: if anyone has contacts with OCCAID, it would be good for them to
find
a way to get IPv6 peering to Verizon Business or give in and get some
transit connectivity.
There's no way to get from OCCAID to places in VZ Business hosting.
My attempts to fix this from the SiXXS and VZ sides weren't able to go
anywhere.
For example:
$ traceroute6 2600:807:320:600::3f55:2416
traceroute to 2600:807:320:600::3f55:2416 (2600:807:320:600::3f55:2416)
from 2001:4830:1100:1fa::2, 30 hops max, 16 byte packets
1 gw-507.bos-01.us.sixxs.net (2001:4830:1100:1fa::1) 2.705 ms 2.736 ms
2.673 ms
2 sixxs-ic-1139-bos.customer.occaid.net (2001:4830:e1:b::2) 2.647 ms
2.742 ms 2.751 ms
3 dcr01-ve402.bstn01.occaid.net (2001:4830:e1:b::1) 2.898 ms 2.939 ms
2.902 ms
4 dcr01-ve402.bstn01.occaid.net (2001:4830:e1:b::1) 2.812 ms !N 2.871
ms !N 2.831 ms !N
On Wed, Oct 21, 2015 at 7:18 PM, Greg Troxel <gdt@lexort.com> wrote:
>
> "Alejandro R. Sede=C3=B1o" <asedeno@mit.edu> writes:
>
> > Sorry, too much abstraction. The VM crashed, and we were switched to a
> > new endpoint. I'm not sure what the state of the crashed VM is.
> >
> > We moved from 216.93.240.245 to 208.118.234.242; OCCAID rebuilt our
> > tunnel on their end, and BGP traffic is flowing once between OCCAID and
> > limekiller.
>
> That's interesting. What's really odd is that I can ping the occaid end
> of limekiller's tunnel which seems to also support sixxs's tunnel. (I
> can also ping Erik's gateway which is not a surprise.) But not
> limekiller's end of the tunnel. I bet the packets are arriving, but
> limekiller does not have a route back to 2001:4830:1100::/48.
>
> ahi root 36 /usr/pkgsrc #> traceroute6 2001:4830:e1:8::1
> traceroute6 to 2001:4830:e1:8::1 (2001:4830:e1:8::1) from
> 2001:4830:1100:249::2, 64 hops max, 12 byte packets
> 1 gw-586.bos-01.us.sixxs.net 10.876 ms 9.171 ms 8.83 ms
> 2 sixxs-ic-1139-bos.customer.occaid.net 8.851 ms 8.748 ms 8.85 ms
> 3 dcr01-tu100.bstn01.occaid.net 8.922 ms 12.915 ms 8.832 ms
> ahi root 37 /usr/pkgsrc #> traceroute6 2001:4830:e1:8::2
> traceroute6 to 2001:4830:e1:8::2 (2001:4830:e1:8::2) from
> 2001:4830:1100:249::2, 64 hops max, 12 byte packets
> 1 gw-586.bos-01.us.sixxs.net 10.455 ms 9.13 ms 6.538 ms
> 2 sixxs-ic-1139-bos.customer.occaid.net 10.902 ms 10.894 ms 8.692 m=
s
> 3 dcr01-ve402.bstn01.occaid.net 8.681 ms 8.833 ms 10.986 ms
> ^C
>
> Anyway, I expect sixxs will resolve this in the next 24h or so and I
> will stop posting traceroutes :-)
>
>
--001a11330cba22b02d0522ca0194
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
<div dir=3D"ltr"><div><div><div><div>Looks like this was fixed around 8pm l=
ast night or a little before.<br><br></div><br>Side-note: if anyone has con=
tacts with OCCAID, it would be good for them to find<br></div>a way to get =
IPv6 peering to Verizon Business or give in and get some transit connectivi=
ty.<br></div>There's no way to get from OCCAID to places in VZ Business=
hosting.<br></div>My attempts to fix this from the SiXXS and VZ sides were=
n't able to go anywhere.<br>For example:<br><br>$ traceroute6 2600:807:=
320:600::3f55:2416<br>traceroute to 2600:807:320:600::3f55:2416 (2600:807:3=
20:600::3f55:2416) from 2001:4830:1100:1fa::2, 30 hops max, 16 byte packets=
<br>=C2=A01=C2=A0 <a href=3D"http://gw-507.bos-01.us.sixxs.net">gw-507.bos-=
01.us.sixxs.net</a> (2001:4830:1100:1fa::1)=C2=A0 2.705 ms=C2=A0 2.736 ms=
=C2=A0 2.673 ms<br>=C2=A02=C2=A0 <a href=3D"http://sixxs-ic-1139-bos.custom=
er.occaid.net">sixxs-ic-1139-bos.customer.occaid.net</a> (2001:4830:e1:b::2=
)=C2=A0 2.647 ms=C2=A0 2.742 ms=C2=A0 2.751 ms<br>=C2=A03=C2=A0 <a href=3D"=
http://dcr01-ve402.bstn01.occaid.net">dcr01-ve402.bstn01.occaid.net</a> (20=
01:4830:e1:b::1)=C2=A0 2.898 ms=C2=A0 2.939 ms=C2=A0 2.902 ms<br>=C2=A04=C2=
=A0 <a href=3D"http://dcr01-ve402.bstn01.occaid.net">dcr01-ve402.bstn01.occ=
aid.net</a> (2001:4830:e1:b::1)=C2=A0 2.812 ms !N=C2=A0 2.871 ms !N=C2=A0 2=
.831 ms !N<br><br><br></div><div class=3D"gmail_extra"><br><div class=3D"gm=
ail_quote">On Wed, Oct 21, 2015 at 7:18 PM, Greg Troxel <span dir=3D"ltr">&=
lt;<a href=3D"mailto:gdt@lexort.com" target=3D"_blank">gdt@lexort.com</a>&g=
t;</span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0=
.8ex;border-left:1px #ccc solid;padding-left:1ex"><span class=3D""><br>
"Alejandro R. Sede=C3=B1o" <<a href=3D"mailto:asedeno@mit.edu"=
>asedeno@mit.edu</a>> writes:<br>
<br>
</span><span class=3D"">> Sorry, too much abstraction. The VM crashed, a=
nd we were switched to a<br>
> new endpoint. I'm not sure what the state of the crashed VM is.<br=
>
><br>
> We moved from 216.93.240.245 to 208.118.234.242; OCCAID rebuilt our<br=
>
> tunnel on their end, and BGP traffic is flowing once between OCCAID an=
d<br>
> limekiller.<br>
<br>
</span>That's interesting.=C2=A0 What's really odd is that I can pi=
ng the occaid end<br>
of limekiller's tunnel which seems to also support sixxs's tunnel.=
=C2=A0 (I<br>
can also ping Erik's gateway which is not a surprise.)=C2=A0 But not<br=
>
limekiller's end of the tunnel.=C2=A0 I bet the packets are arriving, b=
ut<br>
limekiller does not have a route back to 2001:4830:1100::/48.<br>
<br>
ahi root 36 /usr/pkgsrc #> traceroute6 2001:4830:e1:8::1<br>
traceroute6 to 2001:4830:e1:8::1 (2001:4830:e1:8::1) from<br>
2001:4830:1100:249::2, 64 hops max, 12 byte packets<br>
=C2=A01=C2=A0 <a href=3D"http://gw-586.bos-01.us.sixxs.net" rel=3D"noreferr=
er" target=3D"_blank">gw-586.bos-01.us.sixxs.net</a>=C2=A0 10.876 ms=C2=A0 =
9.171 ms=C2=A0 8.83 ms<br>
=C2=A02=C2=A0 <a href=3D"http://sixxs-ic-1139-bos.customer.occaid.net" rel=
=3D"noreferrer" target=3D"_blank">sixxs-ic-1139-bos.customer.occaid.net</a>=
=C2=A0 8.851 ms=C2=A0 8.748 ms=C2=A0 8.85 ms<br>
=C2=A03=C2=A0 <a href=3D"http://dcr01-tu100.bstn01.occaid.net" rel=3D"noref=
errer" target=3D"_blank">dcr01-tu100.bstn01.occaid.net</a>=C2=A0 8.922 ms=
=C2=A0 12.915 ms=C2=A0 8.832 ms<br>
ahi root 37 /usr/pkgsrc #> traceroute6 2001:4830:e1:8::2<br>
traceroute6 to 2001:4830:e1:8::2 (2001:4830:e1:8::2) from<br>
2001:4830:1100:249::2, 64 hops max, 12 byte packets<br>
=C2=A01=C2=A0 <a href=3D"http://gw-586.bos-01.us.sixxs.net" rel=3D"noreferr=
er" target=3D"_blank">gw-586.bos-01.us.sixxs.net</a>=C2=A0 10.455 ms=C2=A0 =
9.13 ms=C2=A0 6.538 ms<br>
=C2=A02=C2=A0 <a href=3D"http://sixxs-ic-1139-bos.customer.occaid.net" rel=
=3D"noreferrer" target=3D"_blank">sixxs-ic-1139-bos.customer.occaid.net</a>=
=C2=A0 10.902 ms=C2=A0 10.894 ms=C2=A0 8.692=C2=A0 ms<br>
=C2=A03=C2=A0 <a href=3D"http://dcr01-ve402.bstn01.occaid.net" rel=3D"noref=
errer" target=3D"_blank">dcr01-ve402.bstn01.occaid.net</a>=C2=A0 8.681 ms=
=C2=A0 8.833 ms=C2=A0 10.986 ms<br>
^C<br>
<br>
Anyway, I expect sixxs will resolve this in the next 24h or so and I<br>
will stop posting traceroutes :-)<br>
<br>
</blockquote></div><br></div>
--001a11330cba22b02d0522ca0194--
