[BBLISA] Amazon EC2 Oddly Rejecting Very Specific IP Addresses

[Dean Anderson]

This is what I see just now:

/usr/sbin/tracepath 75.101.149.255
 1?: [LOCALHOST]     pmtu 1500
 1:  130.105.36.65 (130.105.36.65)                          0.881ms
 2:  av8-qncy-gw1.av8.net (130.105.32.14)                  25.077ms
 3:  qcy1-ar2-f2-0-314.gnaps.net (199.232.42.53)           11.327ms
 4:  ae-0-0-0-030.br1.qcy1.ma.gnaps.net (199.232.44.133)   12.245ms
 5:  so-6-0-0-000.br1.bos1.ma.gnaps.net (199.232.44.9)    asymm  6  
10.730ms
 6:  ge-7-21.car1.Boston1.Level3.net (4.53.48.1)           19.598ms
 7:  ae-2-7.bar1.Boston1.Level3.net (4.69.132.242)        asymm 11  
20.127ms
 8:  ae-0-11.bar2.Boston1.Level3.net (4.69.140.90)        asymm 10  
18.103ms
 9:  ae-8-8.ebr1.NewYork1.Level3.net (4.69.140.98)        asymm  8  
17.907ms
10:  ae-3-3.ebr4.Washington1.Level3.net (4.69.132.93)      42.535ms
11:  ae-74-74.csw2.Washington1.Level3.net (4.69.134.182)  asymm  9  
26.815ms
12:  ae-2-79.edge1.Washington1.Level3.net (4.68.17.80)    asymm 10  
31.657ms
13:  AMAZONCOM.edge1.Washington1.Level3.net (4.79.20.22)  asymm 10  
24.668ms
14:  72.21.199.38 (72.21.199.38)                          asymm 11  
27.141ms
15:  72.21.222.147 (72.21.222.147)                        asymm 12  
30.804ms
16:  no reply
17:  no reply
18:  no reply
19:  no reply
20:  no reply
21:  no reply
22:  no reply
23:  no reply
24:  no reply
25:  no reply
26:  no reply
27:  no reply
28:  no reply
29:  no reply
30:  no reply
31:  no reply
     Too many hops: pmtu 1500


On Tue, 13 Apr 2010, Richard 'Doc' Kinne wrote:

> Folks:
> 
> I'd like to see if anyone has heard of circumstances like this before.
> 
> I have a server on the Amazon EC2 cloud running a website service. This is
> largely working well.
> 
> However I have one customer that cannot get to it from a specific address.
> The IP address of my server is 75.101.149.255. When you do a "whois" on this
> it comes up as:
> 
> OrgName:    Amazon.com, Inc.
> OrgID:      AMAZO-4
> Address:    Amazon Web Services, Elastic Compute Cloud, EC2
> Address:    1200 12th Avenue South
> City:       Seattle
> StateProv:  WA
> PostalCode: 98144
> Country:    US
> 
> which makes perfect sense.
> 
> We originally thought that Amazon might be blocking access to the service to
> specific IP or IP ranges, but based on traceroutes that didn't seem to make
> sense.
> 
> When my customer tries to do a traceroute from his place to my server he
> doesn't even get out of his router:
> 
> tracert 75.101.149.255
> Tracing route to [75.101.149.255] over a maximum of 30 hops:
>   1     1 ms    <1 ms    <1 ms  www.routerlogin.com [10.1.1.1]
>   2     *        *        *     Request timed out.
> 
> I've never seen anything like that before. I can understand things timing
> out when you get to the Amazon area, but timing out before you even get into
> the Net proper? That doesn't make sense to me. Everything else seems to work
> properly from his location from what he's telling me.
> 
> There is a part of me that thinks there may be something wrong somehow with
> my customer's address. When I do a "whois" on the customer's address it
> comes back as being owned by IANA, which doesn't seem right at all. Also
> when I try a traceroute to his address *I* don't get past my router in two
> totally separate locations (work, that has one ISP, and home, which has a
> very different ISP).
> 
> I've never quite seen anything act like this before and I'm not quite sure
> how to puzzle it out.
> 
> Does anyone have any thoughts?
> 

-- 
Av8 Internet   Prepared to pay a premium for better service?
www.av8.net         faster, more reliable, better service
617 256 5494