maas-devel team mailing list archive

[Julian Edwards <julian.edwards@xxxxxxxxxxxxx>]

Thank you both for the replies.

On Thursday 24 May 2012 11:44:32 Francis J. Lacoste wrote:
> Hi Julian,
> 
> I agree with Michael here, I think we should look into defaulting based
> on auto-discovered information rather than creating the concept of
> profile or group which would still mean manual data entry.

I think you both misunderstood slightly here.  I agree with what you are 
saying but you missed the point - you can't discover power parameters like 
that (AFAIK).  What I am talking about here is rack/subnet-specific things 
outside of the node's hardware.

> But we should probably leave this for after we introduce the hardware
> database (which we want for better Juju constraints).

For that sort of defaulting, I completely agree.

> On 12-05-24 07:20 AM, Michael Hughes wrote:
> > Hi Julian,
> > 
> > Agreed that a 'default' template would make scaling much faster, but
> > rather than assuming one would set up multiple subnets for different
> > hardware types I'd recommend looking into DMI queries to automatically
> > determine machine type.

I'd not want to mandate a subnet for each harware type.

> > The Capone plugin over at the FOG project is an excellent example of
> > how to implement this:
> > http://www.fogproject.org/wiki/index.php?title=Plugins:_Capone#Using_Capon
> > e
> > 
> > and
> > http://www.fogproject.org/wiki/index.php?title=Capone

Thanks for the pointers.

J