Add support for latency-based routing on AWS

[kaazoo]

Route53 on AWS supports "latency-based routing" for DNS records. You can have multiple DNS records for the same hostname, having different ALIAS to regional ELBs.
This is usually the preferred way of having regional failover and low latency for multi-region deployments.

[cristim]

I'd like to implement this, unless someone else is working on it.

[cristim]

@hjacobs please have a look at 4cd8257 and let me know if you have any concerns. Should I maybe make this configurable with an additional annotation?

The current change passes the unit tests but I didn't yet run it against an actual environment.

[hjacobs]

@cristim looking through your mentioned commit, I don't really know what it does: would it enable latency based routing "always"? Is there also an AWS cost factor?

[cristim]

@hjacobs indeed, I just had a look and it turns out the cost per DNS request is 50% more than the simple records: https://aws.amazon.com/route53/pricing/

From my experience it's still often negligible when compared to the other infrastructure costs, but I guess this makes sense to be an opt-in feature since there are many people out there where it doesn't give any additional value so they shouldn't pay an extra price for it.

I'll try to make this configurable and extensible to other Route53 record types that may be supported later (geolocation, weighting, etc.).

[meringu]

+💯

[alfredkrohmer]

@cristim Any updates on this?

[cristim]

@devkid I stopped working on this, and I am now unable to resume for the next couple of months. If you are interested feel free to pick up from where I left it.

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[alfredkrohmer]

I'm working on this now.

[alfredkrohmer]

/remove-lifecycle stale

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[aimanparvaiz]

would be great to have this.

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten

[fejta-bot]

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

[k8s-ci-robot]

@fejta-bot: Closing this issue.

In response to this:

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[cristim]

@AlaaMansourTW apparently yes, see #1008.

[ghost]

@cristim
Yes, managed to get it done with - --txt-prefix and having a unique identifier

[jnicholls]

I think this would be an excellent feature to be expressible via kube semantics, like other cloud-provider-aws and external-dns features are. Is this issue on anyone's radar at all?