[as112-ops] AS112 node via AS5408

[Scott Mace]

Internet2 and NLR have decided that it is best to filter AS112 announcements
at their edges and investigate running their own AS112 instances.

	Scott

William F. Maton Sotomayor wrote:
> On Sun, 18 Nov 2007, Scott Mace wrote:
> 
>> I'm not seeing the prefix via your 6509 2884 path, still only the
>> geant2 path.  I tested that server coming from 5408 again last
>> night and it was still giving timeouts.
> 
> Hi Scott.   Somewhat of a postscript, but I believe that the errant node 
> might be gone. I never did get a response from a human at GRNet other 
> than the usual automated message, so perhaps they've realized there was 
> a problem.
> 
>>
>>     Scott
>>
>> William F. Maton Sotomayor wrote:
>>> On Fri, 16 Nov 2007, Scott Mace wrote:
>>>
>>>> Oddly enough at some point in the past (who knows how long) an AS112
>>>> node started being announced to research networks in the US.   AS11537,
>>>> for example:   302 4557 14085 11537 20965 5408 112
>>>>
>>>> The only problem is that many of the end sites connected to 11537
>>>> prefer all paths via 11537 blindly over other paths.
>>>> So as a result this particular AS112 node might get more than it 
>>>> expected.
>>>> It is currently responding with timeouts.
>>>
>>> Interesting, I wonder if we have a contact here for GRNet.  In any 
>>> case, I'll see if I can contact them about.
>>>
>>> FWIW, my AS112 node is back online, and may help with timeouts.  
>>> Testing appreciated.  I'm coming in from behind AS path 6509 2884 and 
>>> you can give this a shot:
>>>
>>> dig @prisoner.iana.org hostname.as112.net txt +short +norec
>>>
>>> This might come back with:
>>>
>>> "Federal GigaPOP" "Ottawa, Canada"
>>> "See http://as112.net/ for more information."
>>>
>>> Provided we're 'closer' to you than others.
>>>
>>> wfms
>>>
>>> P.S. I have asked AS6509 to peer direct with them instead of the 
>>> route servers.
>>>
>>>
>>
> 
> 
> wfms
> 
>