[dns-operations] anycast ops willing to review a IETF draft?

[Jeff Westhead]

This is coming late but a few comments:

IPv6 traffic may take different network paths which can result in higher latencies. When we enabled IPv6 in Azure DNS it seemed like IPv6 has more limited transit / peering than IPv4 so interestingly you may be worse-off using IPv6 for the moment.

Regarding R1, consider advertising different subsets of Anycast IPs from different locations. This may minimize impact of network issues in a particular location. For example if a grey router failure causes packet loss to a particular PoP advertising one set of IPs, recursive resolvers may be able to prioritize an IP in another subset, routing queries to another nearby PoP which advertises the differing IPs. This can decrease the risk of end users being impacted by networking issues. 

DDoS scrubbing to identify and drop malicious traffic as close to attackers, or as close to the edge of the operator's network, as is feasible is advisable.

Any recommendations with respect to RFC 8482?

-----Original Message-----
From: dns-operations <dns-operations-bounces at dns-oarc.net> On Behalf Of Giovane Moura
Sent: Monday, February 25, 2019 2:07 AM
To: dns-operations at dns-oarc.net
Subject: [dns-operations] anycast ops willing to review a IETF draft?

Folks,

So we have a recommendations draft that we will present at the next DNSOP meeting at the IETF.

We've been asked to contact Anycast operators to review our draft, so we can consider their input to make the draft better.

Any anycast ops here willing to review the draft? Any feedback is really appreciated.

here are the details, and thanks a lot in advance.

/giovane


########################################################################

Draft:

https://www.ietf.org/id/draft-moura-dnsop-authoritative-recommendations-01.txt

Major changes from -00:

  * Fixed scope:  folks pointed that our recommendations (such as use
anycast) were not for everyone. We added text and focus on "large operators",  "those with a significant global user  population"
  * We have enumerated 00 reviewers' comments in [0].  Each comment is labeled as [RF:AB01], where
     *RF: review-fix
     *AB: reviewer's initials
     *01:reviewer AB 01 comment
  * We then annotated our version -01 with such labels, indicating
*where* we fixed it. These labels will be later removed.
  * The most 'complicated' comments were turned into "Issues" on Github; we can see how we addressed them in [1].


[0]
https://github.com/gmmoura/draft-moura-dnsop-authoritative-recommendations/blob/master/reviews/reviews-dnsop.md

[1]
https://github.com/gmmoura/draft-moura-dnsop-authoritative-recommendations/issues


_______________________________________________
dns-operations mailing list
dns-operations at lists.dns-oarc.net
https://lists.dns-oarc.net/mailman/listinfo/dns-operations
dns-operations mailing list
https://lists.dns-oarc.net/mailman/listinfo/dns-operations