http://defect.opensolaris.org/bz/show_bug.cgi?id=9457





--- Comment #4 from Lizhong Li <lizhong.li at sun.com>  2009-06-15 18:20:51 ---
(In reply to comment #3)
> (In reply to comment #2)
> > I was just thinking that it's better if nwamd could just enable the ncus
> > related with the plumbed/up links, and leave other ncus related with
> > unplumbed/down links disabled,
> 
> I'm not sure what you mean here. All the links were plumbed and up.
> Do you mean that we should unplumb links we can't get addresses for? 
> If so, I disagree - it may be the case we have a slow-responding DHCP 
> server. We have a timeout mechanism that will cause us to give up
> if all NCUs in a shared priority group fail to find an address, but
> that's not the case in your configuration - nge0 succeeded.
> 
> > this may give the users the consistent feeling
> > with non-nwam cases. And this may make the output of 'ifconfig -a' more 
> > clean.
> 
> I don't think there's a way to minimize plumbing of interfaces
> in a configuration like the one you have, short of modifying
> the User NCP to disable the links you know can't get a DHCP address.
> In a way, that's what your non-NWAM configuration does - it says
> "start DHCP on nge0 only".
Yes, what I mean is that I was assuming the same behavior of non-NWAM and nwam
configuration, while I think perhaps it's just my misunderstanding of the
mechanism of nwam which is really different from physical:default. 
What I supposed is , the system boots with physical:default, only bge0 is
plumbed and gets 'UP', other links like 'bge1 bge2' are unplumbed, so I'd like
to see the nwam could keep this state when nwam replace physical:default later,
i.e., keep 'bge1 bge2' like before.

-- 
Configure bugmail: http://defect.opensolaris.org/bz/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.
You are the assignee for the bug.

Reply via email to