prep for cutting a release, target 2019-01-13

[Hal Murray]

> I want to mention that the bug (and regression w.r.t. NTP classic) I reported
> that effectively drops configured servers by changing to poll interval to a
> high value is still present, even though Eric declared it a top priority ove
> one year and four months ago.

Is the problem that it doesn't recover, or that it doesn't recover as fast as 
it should?  I see the polling interval stretched out when a server isn't 
responding, but I also see it returned to normal the next time I look long 
after the server recovers.  I haven't looked carefully at things while it is 
changing.

What's the Issue number?

-- 
These are my opinions.  I hate spam.