[outages] Google 8.8.8.8 Resolution of Route53 domains

[Ryan McGinnis]

I suspect this was related to this issue (via ycombinator hacker news):

https://doublepulsar.com/hijack-of-amazons-internet-domain-service-used-to-
reroute-web-traffic-for-two-hours-unnoticed-3a6f0dda6a6f


On Tue, Apr 24, 2018 at 8:51 AM, Zach Hanna via Outages <outages at outages.org
> wrote:

> Resolved here too..
>
> On Tue, Apr 24, 2018 at 7:30 AM Phil Lavin via Outages <
> outages at outages.org> wrote:
>
>> Those prefixes had been withdrawn now – traffic is flowing correctly
>> again for us. If that was the cause, I suspect things are back to rights
>> for everyone now?
>>
>> One wonders why HE doesn’t apply filters on a peer with 20 legit prefixes…
>>
>>
>>
>> *From:* Outages <outages-bounces at outages.org> *On Behalf Of *Joseph B
>> via Outages
>> *Sent:* 24 April 2018 13:56
>>
>>
>> *To:* outages at outages.org
>> *Subject:* Re: [outages] Google 8.8.8.8 Resolution of Route53 domains
>>
>>
>>
>>
>>
>> Tue Apr 24 11:05:41 UTC onwards one of Hurricane Electric's peers AS10297 started
>> advertising the following subnets via HE.
>>
>>
>>
>> 205.251.192.0
>>
>> 205.251.193.0
>>
>> 205.251.195.0
>>
>> 205.251.197.0
>>
>> 205.251.199.0
>>
>>
>>
>> These are all Amazon subnets, usually originated as part of /23s and
>> seemingly host a fair bit of AWS Route53.
>>
>>
>>
>> If you (or your DNS resolver) are a HE transit customer you will be
>> impacted the most.
>>
>>
>>
>> Cheers,
>>
>>
>>
>> Joseph
>>
>>
>>
>>
>>
>> On Tue, Apr 24, 2018, at 9:50 PM, Phil Lavin via Outages wrote:
>>
>> This doesn’t feel right, though I’ll admit I’ve never checked before. Our
>> only route to ns-163.awsdns-20.com (205.251.192.163) is through HE:
>>
>> inet.0: 757581 destinations, 2107440 routes (757301 active, 0 holddown,
>> 522 hidden)
>>
>> + = Active Route, - = Last Active, * = Both
>>
>> 205.251.192.0/24   *[BGP/170] 01:12:08, localpref 70
>>
>>                       AS path: 6939 10297 I, validation-state: unverified
>>
>>                     > to 216.66.90.21 via ge-1/0/5.0
>>
>> AS10297 is eNET inc. Is this expected?
>>
>>
>>
>> *From:* Outages <outages-bounces at outages.org> *On Behalf Of *Phil Lavin
>> via Outages
>> *Sent:* 24 April 2018 13:04
>> *To:* outages at outages.org
>> *Subject:* Re: [outages] Google 8.8.8.8 Resolution of Route53 domains
>>
>>
>>
>> Looks more specific to AWS than it does to Google+AWS. Can’t resolve
>> against some of AWS’s NS directly:
>>
>> phil at phil-debian:~$ dig cloudcall.com IN A @ns-163.awsdns-20.com
>>
>> ; <<>> DiG 9.10.3-P4-Debian <<>> cloudcall.com IN A @ns-163.awsdns-20.com
>>
>> ;; global options: +cmd
>>
>> ;; connection timed out; no servers could be reached
>>
>>
>>
>> *From:* Outages <outages-bounces at outages.org> *On Behalf Of *Phil Lavin
>> via Outages
>> *Sent:* 24 April 2018 12:56
>> *To:* outages at outages.org
>> *Subject:* Re: [outages] Google 8.8.8.8 Resolution of Route53 domains
>>
>>
>>
>> Yeh. Still digging into it.
>>
>>
>>
>> *From:* Outages <outages-bounces at outages.org> *On Behalf Of *Zach Hanna
>> via Outages
>> *Sent:* 24 April 2018 12:54
>> *To:* outages at outages.org
>> *Subject:* [outages] Google 8.8.8.8 Resolution of Route53 domains
>>
>>
>>
>> Anyone else seeing SERVFAIL for route53-hosted domains trying to resolve
>> with Google DNS?
>>
>> *_______________________________________________*
>>
>> Outages mailing list
>>
>> Outages at outages.org
>>
>> https://puck.nether.net/mailman/listinfo/outages
>>
>>
>> _______________________________________________
>> Outages mailing list
>> Outages at outages.org
>> https://puck.nether.net/mailman/listinfo/outages
>>
>
> _______________________________________________
> Outages mailing list
> Outages at outages.org
> https://puck.nether.net/mailman/listinfo/outages
>
>


-- 
-Ryan McGinnis
Platte Valley Communications
308-237-9512
PGP: 62E39BC1
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/outages/attachments/20180424/d355f059/attachment.htm>