[175638] in North American Network Operators' Group

home help back first fref pref prev next nref lref last post

Re: pay.gov and IPv6

daemon@ATHENA.MIT.EDU (Matthew Kaufman)
Sun Oct 26 21:39:53 2014

X-Original-To: nanog@nanog.org
In-Reply-To: <20141026211643.822F3228311D@rock.dv.isc.org>
From: Matthew Kaufman <matthew@matthew.at>
Date: Sun, 26 Oct 2014 18:39:44 -0700
To: Mark Andrews <marka@isc.org>
Cc: NANOG list <nanog@nanog.org>
Errors-To: nanog-bounces@nanog.org

This is why I need to pull logs the next time I need to pay the FCC. There a=
re several rounds of redirects involved from clicking the payment button on t=
he FCC site to the final landing at pay.gov, and one of the last steps never=
 connects if IPv6 is enabled.

Matthew Kaufman

(Sent from my iPhone)

> On Oct 26, 2014, at 2:16 PM, Mark Andrews <marka@isc.org> wrote:
>=20
>=20
> In message <CAFG21ohZ6MV6Tef_sWuwV6kmAZmHQ3nFRLq-FkdU38g=3DvL3nnQ@mail.gma=
il.com>
> , Todd Lyons writes:
>> On Sat, Oct 25, 2014 at 10:26 AM, Matthew Kaufman <matthew@matthew.at> wr=
ote:
>>>>=20
>>>> Random IPv6 complaint of the day: redirects from FCC.gov to pay.gov fai=
l
>>>> when clients have IPv6 enabled. Work fine if IPv6 is off. One more set o=
f
>>> Still broken, 7 months later. And again, I was too busy trying to pay to=
 tr
>> y
>>> to pull a full set of logs. But if you do something on the FCC site that=

>>> requires payment, the redirection flow dies halfway through if you're co=
min
>> g
>>> from IPv6 and works fine if you turn it off... so yet another computer i=
n
>>> the house has IPv6 disabled until manually turned back on.
>>=20
>> FWIW, eftps.gov is also unreachable via ipv6. I tried all of miredo,
>> and my home Sixxs tunnel, and a HE tunnel from somewhere else.  I used
>> the 4or6 plugin to temporarily disable ipv6 and both sites loaded
>> straight away.
>=20
> If a site is unreachable your client should switch to IPv4 unless
> a IPv6 literal has been used.
>=20
> If your client take ages to switch over report a bug to the client
> vendor.
>=20
> It should not take ages to switch between multiple server addresses.
> IPv4 + IPv6 is just a example of multiple server addresses.
>=20
>> eftps.gov and pay.gov appear to be managed separately since both their
>> ipv4 and ipv6 netblocks are not in the same netblocks, and my path to
>> them is not the same:
>>=20
>> eftps.gov has IPv6 address 2620:10f:400e:a::13
>> mtr to eftps.gov via Sixxs:
>> Host                                          Loss%   Snt   Last
>> Avg  Best  Wrst StDev
>> 1. 2604:8800:100:82bc:ddcb:ae62:e3da:c91f      0.0%    16    0.9
>> 0.9   0.9   1.6   0.2
>> 2. gw-701.chi-03.us.sixxs.net                  0.0%    16   71.6
>> 72.4  68.6  78.3   2.4
>> 3. uschi03.sixxs.net                           0.0%    16   70.2
>> 71.8  69.2  78.8   2.4
>> 4. 2620:0:6b0:a::1                             0.0%    15   67.3
>> 73.3  67.3  79.7   3.2
>> 5. tge3-1.fr3.ord4.ipv6.llnw.net               0.0%    15   73.6
>> 75.4  70.1  85.4   4.9
>> 6. ve8.fr3.ord.ipv6.llnw.net                   0.0%    15   73.5
>> 79.7  72.9  90.4   5.7
>> 7. 2600:805:41f::5                             0.0%    15  104.4
>> 81.9  74.2 104.4   9.0
>> 8. 2600:806::12                                0.0%    15  105.2
>> 104.0 100.6 109.8   2.9
>> 9. 2600:806:12f::2e                            0.0%    15  134.5
>> 135.7 131.4 147.4   4.1
>> 10. 2620:10f:400e:1::4004                       0.0%    15  161.5
>> 145.9 131.5 163.8   9.9
>> 11. ???
>>=20
>> pay.gov has IPv6 address 2605:3100:fffd:100::15
>> mtr to pay.gov via Sixxs:
>> Host                                          Loss%   Snt   Last
>> Avg  Best  Wrst StDev
>> 1. 2604:8800:100:82bc:ddcb:ae62:e3da:c91f      0.0%    11    0.9
>> 0.9   0.7   1.1   0.1
>> 2. gw-701.chi-03.us.sixxs.net                  0.0%    11   70.8
>> 70.9  67.0  74.4   2.2
>> 3. uschi03.sixxs.net                           0.0%    11   73.7
>> 73.7  69.8  90.1   5.6
>> 4. 2620:0:6b0:a::1                             0.0%    11   70.6
>> 73.7  70.4  86.2   5.0
>> 5. tge3-1.fr3.ord4.ipv6.llnw.net               0.0%    11   72.4
>> 75.6  71.5  82.6   3.2
>> 6. ve8.fr3.ord.ipv6.llnw.net                   0.0%    11   76.1
>> 79.3  74.7  87.9   4.0
>> 7. tge32-3.fr3.dal.ipv6.llnw.net               0.0%    11   99.1
>> 100.1  96.4 106.2   2.7
>> 8. sl-st30-dal-te0-14-0-1.v6.sprintlink.net    0.0%    11   98.2
>> 102.0  98.2 111.0   4.4
>> 9. sl-crs1-fw-be40.v6.sprintlink.net           0.0%    11   99.5
>> 100.5  96.2 105.5   2.5
>> 10. sl-gw38-fw-po0-0.v6.sprintlink.net          0.0%    11   96.4
>> 98.8  96.4 105.1   2.6
>> 11. 2600:4:2000:4::9                            0.0%    11  100.2
>> 102.0  99.0 107.0   2.7
>> 12. ???
>>=20
>> I was hoping an eftps.gov or pay.gov employee was casting an eye this
>> way, but it doesn't look like anybody from there is subscribed to
>> NANOG.
>>=20
>> ...Todd
>> --=20
>> The total budget at all receivers for solving senders' problems is $0.
>> If you want them to accept your mail and manage it the way you want,
>> send it the way the spec says to. --John Levine
> --=20
> Mark Andrews, ISC
> 1 Seymour St., Dundas Valley, NSW 2117, Australia
> PHONE: +61 2 9871 4742                 INTERNET: marka@isc.org

home help back first fref pref prev next nref lref last post