On Tue, Mar 28, 2017 at 6:20 AM, Pieter Lexis <[email protected]> wrote:
> On Mon, 27 Mar 2017 18:19:09 -0400
> Jared Mauch <[email protected]> wrote:
>
>> I will note there are other implementations out there as well, such as in 
>> unbound.  serve-expired configuration directive is available there as well.
>
> I feel that the authors should attempt to describe the goal of the algorithm 
> and suggest possible limits and describe pitfalls rather than describing the 
> exact algorithm to use.


Noted.
As one of the authors of both this, and an earlier document with a
different algorithm ( draft-wkumari-dnsop-ttl-stretching ), I think
that the actual algorithm specified is a secondary consideration --
first we need to agree on if the concept / idea is good for general
use.

Assuming that it is a good idea to standardize (perhaps only for
certain use-cases / environments)...

This document has an algorithm which has been deployed, and tuned and
fettled, and works nicely -- however, the environment in which is
being used in may different attributes to the general case, so...


It is entirely possible that:
A: the described algorithm and parameters are perfect for general use, or
B: the described algorithm with different parameters is perfect for general use
C: we need a different algorithm for general use
D: we could describe multiple algorithms, and implementors can select
E: we can just describe what the expected behavior is, and
implementors can determine how to do this.
F: something else

I personally think that B is reasonable -- the algorithm seems to
work, but different (or adjustable) timers might be better for general
recursive servers. But, I'm happy to be wrong -- if the WG adopts the
document it owns it and can choose any of the above...


> This will allow implementers to come up with new and innovative algorithms 
> based on continued measurements.
> I do support an adoption of such a draft in the WG.

Thank you.
W


>
> Best regards,
>
> Pieter
>
> --
> Pieter Lexis
> PowerDNS.COM BV -- https://www.powerdns.com
>
> _______________________________________________
> DNSOP mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/dnsop



-- 
I don't think the execution is relevant when it was obviously a bad
idea in the first place.
This is like putting rabid weasels in your pants, and later expressing
regret at having chosen those particular rabid weasels and that pair
of pants.
   ---maf

_______________________________________________
DNSOP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dnsop

Reply via email to