OMAPI "key conflict"

[Allie Hopkins]

Not quite.

If you have a host declared with a fixed IP, then it will receive that
number IF viable for that subnet.  Otherwise, it will simply grab one
from the dynamic range.

The multiple addresses you are trying to create, I assume, are for two
different networks.  It will pick that one that works.  If none works,
then it picks from the range statement of the correct pool.

Don't create the host three times.  Create it either twice (once for
each IP) or only once with both IPs listed together.  The third host
declaration with just the mac is not needed if you have it defined any
other way.

On Mon, 2007-09-10 at 21:59 +0100, Phil Mayers wrote:
> > So, at the moment all my hosts are declared with one "host" stanza for
> > the mac alone, and zero or more for each IP (there's no particular
> > reason for this - it's just the way the code has always generated them,
> > and it pre-dates me) like so:
> > 
> > host 11-22-33-44-55-66 {
> >   hardware ethernet 11:22:33:44:55:66;
> > }
> > 
> > host 192.168.0.10 {
> >   hardware ethernet 11:22:33:44:55:66;
> >   fixed-address 192.168.0.10;
> > }
> > 
> > host 192.168.1.10 {
> >   hardware ethernet 11:22:33:44:55:66;
> >   fixed-address 192.168.1.10;
> > }
> > 
> > If instead I move to:
> > 
> > host 11-22-33-44-55-66 {
> >   hardware ethernet 11:22:33:44:55:66;
> >   fixed-address 192.168.0.10, 192.168.1.10;
> > }
> > 
> > ...then I can be sure that:
> > 
> >  1. the behaviour of the latter setup is exactly equivalent to the
> > former, without even considering OMAPI
> 
> Bah. It's not is it?
> 
> """The host declarations will only match a client if one of their
> fixed-address statements is viable on the subnet (or shared network)
> where the client is attached.  Conversely, for a host declaration to
> match a client being allocated a dynamic address, it must not have
> any fixed-address statements.  You may therefore need a mixture of
> host declarations for any given client...some having fixed-address
> statements, others without."""
> 
> So if a host initially has no fixed IPs, I'll start with:
> 
> host foo { hardware ethernet $MAC; }
> 
> ...and if someone updates it in my database, it will be impossible using
> OMAPI to either add the fixed-address statements to the host (since that
> will stop it picking up dynamic addresses) or create a 2nd host (since
> the MAC address conflicts).
> 
> How annoying.
> 
>