Interesting Email Server Setup Article from Redhat

[David Bandel]

On 12/7/06, Matthew Carpenter <mcarpenter at intelguardians.com> wrote:
> On Thursday 07 December 2006 15:06, David Bandel wrote:
> > On 12/6/06, James McDonald <james at jamesmcdonald.id.au> wrote:
> > > http://www.redhat.com/magazine/025nov06/features/email/index.html#parttwo
> > >?sc_cid=bcm_edmsept_007
> >
> > <rant>
> > Why do folks insist on using the word "propagate" to talk about DNS?
> > DNS is _passive_ not _active_.  DNS doesn't "propagate" anything.  In
> > DNS, caches time out and when they do, the updated information is
> > retreived.  But nobody else's DNS server contacts mine to propagate
> > (broadcast, circulate, transmit) anything.  Even my master DNS server
> > doesn't broadcast changes to the slaves, it only signals the slaves
> > that something has changed.  The slaves then request the information
> > out of cycle, but the master doesn't just broadcast it.
> > </rant>
>
> Because it's bite-sized.  One word, even if only effectively close to
> accurate, sums up what you spent a paragraph describing, and which most
> people wouldn't be able to follow.

... wait for DNS to propagate ...
... wait for the DNS caches to timeout ...

This isn't like calling a chicken a fowl, it's like calling a chicken a carrot.

>
> I've even taken to using the "propagation" term, since it's quick and
> basically accurate for all intents and purposes.  If I'm speaking to someone

Not even basically accurate.  No broadcasts or other spontaneous
transmissions occur.  Caches timeout.

> who *needs* to understand the difference, they generally do... or then I
> explain it.  But a manager doesn't need to understand the pull mechanism...
> just the fact that the data now lives elsewhere for a certain timeperiod.

And the last sentence makes no sense at all.  What does DNS have to do
with data living anywhere?

David A. Bandel
-- 
Focus on the dream, not the competition.
            - Nemesis Air Racing Team motto