Spec: Update the spec for federation about dns HOT 10 CLOSED

@nikhiljindal One thing to keep in mind, increases in ndots can have a dramatic impact on performance.

When a name has fewer dots than ndots, external names result in 10 DNS lookups/name. Modulo negative caching, this can easily result in DNS server meltdown.

from dns.

The federation records need to be totally revamped - last I looked at them they were ambiguous. @csbell @madhusudancs

from dns.

@thockin The best reference outside of code that I could find is located here. By revamp, do you mean merging the current federation behavior into the spec here or is there more involved?

from dns.

from dns.

Given federation is still in its early stages, it would be great to get ahead of this issue

from dns.

Oh wow! Did not know about the existence of this repo and was hence missing the notifications.

We proposed something like mysvc.myns.myfed.fed.u7s.io as an alternative,
but we need a plan to migrate people. Madhu might recall more details.

This and in addition we had the literal svc somewhere in there, IIRC. Something like mysvc.myns.svc.myfed.fed. We still need to come up with a concrete migration plan.

There is also this doc that Quinton wrote - https://docs.google.com/document/d/1tSTgqSy7iR8cQx2z-bbqaYVITrraDQmV_aE43PuRPoI/edit#heading=h.ys8jzm2jzab9

from dns.

Also ref kubernetes/kubernetes#27969

We proposed something like mysvc.myns.myfed.fed.u7s.io as an alternative,

We had also discussed supporting DNS names for ingress. So fsvc and fing should be better than fed.
Also, with that proposal, we will need more than 6 dots for stateful sets (extra dot for federation name). Is that acceptable?

Given federation is still in its early stages, it would be great to get ahead of this issue

+1.

from dns.

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.

Prevent issues from auto-closing with an /lifecycle frozen comment.

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

from dns.

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
/remove-lifecycle stale

from dns.

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

from dns.