Question on address configuration

Robert Elz kre@munnari.OZ.AU
Thu, 17 Jan 2002 20:21:52 +0700


    Date:        Wed, 16 Jan 2002 07:13:50 -0800
    From:        Nick Sayer <nsayer@quack.kfu.com>
    Message-ID:  <3C4598AE.9040700@quack.kfu.com>

  | Everything you said is true, but I would add the small postscript that 
  | in 99% of the cases where you would find it truly desirable to set up a 
  | well known address,

I wasn't really talking about well known addresses necessarily, just
addresses where I choose the suffix, rather than the system picking its
own.

  | you would be better off making that address 
  | site-local rather than based on the (routable) prefix.

Site local, global, it makes no difference.  The prefix still comes
from the router (I don't want to configure my subnet number any more
than I do my ISP provided prefix - those the node should dynamically learn).

  | Your arguments 
  | about an SMTP server's DNS TTL can be made about any DNS records -- if 
  | the TTLs are causing you heartburn you should lower them.

Of course.   But lowering them affects cache effectiveness, and since
the address is mostly stable, there's no need to do that.   If there's
to be a planned change, then sure, I'll reduce the TTL.   But for an
unexpected change, it is far easier just to keep the same address as I
had before.  That way the TTL can remain large enough to be sensible
most of the time.   Otherwise I'd have to keep it at around 5 or 10
minutes - because predicting when an ethernet card (or whole system) will
die, and need to be replaced with no warning, is beyond my abilities.

  | People 
  | typically want to configure manual addresses so that they don't have to 
  | keep rewriting resolv.conf files (or equiv).

Please stop deciding you know why I want to do what I want to do, and
then providing an alternate means to achieve that.   That's not a very
useful method of discussion.

kre