On 05/18/10 04:52 AM, James Carlson wrote:
Darren J Moffat wrote:
On 17/05/2010 22:42, Erik Nordmark wrote:
Thus replacing network/physical:default with a static NWAM profile would
introduce a new failure mode; if the cable is unplugged the server can
no longer talk it itself, and server applications that bind to the local
IP addresses of the system would fail - since the addresses wouldn't be
configured when the link is down.
Hence we can't use NWAM for this right now.
Thanks for that explanation, that makes a lot more sense as the reason
why NWAM can't be used than wither or not the NWAM configuration of a
static profile is in an SMF xml fragment or not.
It does make more sense, but how is it relevant for system install?
Surely you don't want to have (and can't have) your network interfaces
stuck "down" while attempting to do a network-based install, right?
Erik's description of this problem would be a reason why the current
NWAM would be unsuitable for _installed_ server systems (and is perhaps
a bug), but I don't think it addresses this _installer_ project directly.
I think that one of the underlying arguments in favor of this project is
that since this is a one-shot deal (used only during install time, not
in any subsequent post-install ordinary operation), it doesn't have to
look like anything else, nor include options for all of the current
networking features, nor follow any of the rest of the design in this
area. It's an exception.
The plan is for this project to handle the initial requirements from the
install team, and be useful as a stepping stone as we 1) make the
configuration post install more feature-rich and 2) move the network
configuration into the SMF.
I realize the ARC can't review that, because it isn't done. Hence it is
hard for the ARC to add value in this case.
Even given that situation, I still think the project is misguided. A
better approach would be to help the NWAM project do what's needed
rather than building a bypass out of bailing wire. It seems a shame to
miss the opportunity to have these groups working together rather than
at cross-purposes.
I don't understand how speculating on the organizational structure at
Oracle is part of the ARC process.
FWIW There is one organization working on network configuration. That
includes folks that have previously worked on NWAM.
There is a plan with a set of projects to incrementally improve the
network configuration. I hope we can get those done as soon as possible,
but there is significant work to be done as we complete the transition
to ipadm, move away from /etc/hostname files and towards SMF, and
integrate (NWAM) profiles into the base of the system.
Erik
_______________________________________________
opensolaris-arc mailing list
opensolaris-arc@opensolaris.org
