Re: How many folks are actually using dynamic IP address assignment?

["Michael J. Lewis" <[e-mail]>]

At Chevron, Our goal is to utilize dynamic addressing but because of the
missing 
DNS interface, we've amended that goal in the interim to utilize static
or DHCP
manual addressing for servers.  This policy has generally worked well
particularly
at smaller sites (100 or fewer devices).  The larger sites have had a
bit more
difficulty but these problems are primarily a combination of dynamic
addressing
and server implementation rather than dynamic addressing itself.

We do recognize that dynamic addressing currently is a two-edged sword
and have
undertaken great efforts to document potential problems.  We've given
several
presentations internally to try and educate local support about the
potential
problems.  The absence of the dynamic DNS also is requiring us to
maintain local
distribution/support of IP addresses rather than have this function
centralized
as part of LAN support.  

Anyone wishing further details on our implementation can Email me
directly.

Mark Boolootian wrote:
> 
> I'm curious as to how many sites out there are actually assigning IP addresses
> dynamically.  From some of the comments I've seen on the ISC DHCP server
> mailing list, it sounds as if many sites are avoiding dynamic assignment, and
> opting for nailed up addresses handed out via DHCP.
> 
> Perhaps this is done for accountability purposes?  Or perhaps this is a
> result of the problematic interactions between DNS and DHCP?
> 
> Have those of you running with dynamic addressing encountered any especially
> pernicious problems?  Anyone run into the problem of exhausting the available
> pool of dynamic addresses - how did you discover this problem?  Just allowing
> people to plug in and turn on would seem to necessitate some kind of monitor
> to watch for this sort of thing.
> 
> Any and all comments in relation to this would be greatly appreciated.
> 
> regards,
> mb