Dan Smith wrote:
KR> DefineSystem() doesn't check to see if the guest already
exists.
Hmm, I think it used to because I remember having to undefine a guest
while testing it.
KR> In create_system(), before calling get_reference_domain(), there
KR> should be such a check. If the domain exists, we should get the
KR> domain struct and then update the struct accordingly.
That seems like a different issue to me than the one that this patch
addresses.
That's a fair point. I can follow up with it after this patch goes in.
--
Kaitlin Rupert
IBM Linux Technology Center
kaitlin(a)linux.vnet.ibm.com