http://defect.opensolaris.org/bz/show_bug.cgi?id=10683



--- Comment #5 from amaguire <alan.maguire at sun.com> 2009-08-20 13:26:11 UTC 
---
(In reply to comment #4)
> When doing state transitions, we currently don't care about what the current
> state is.  When generating state events, we only look at the event that is
> causing the state change (say routing socket message of RTM_NEWADDR)and
> generate the state change event.  When actually changing the state, we don't
> look at the existing state.  So, an IP NCU can be disabled or offline or
> uninitialized, and if for some reason we receive an RTM_NEWADDR we move to
> online, if up regardless of the previous state.
> 
> Bug 10456 is also somehow related to this.  We transition state of the 
> wireless
> NCU without considering that it is already in offline state when the wired NCU
> came online.

I've tried to tackle some of this with my last push - IF_STATE events are now
ignored if the NCU is not in offline* or online, and wireless is a bit more
careful too - I'd noticed that a reinit (which happens for wireless interfaces
on refresh/rescan) will throw away the fact that the link may have been
disabled for example. It'd be worth trying these actions again with the latest
build containing those changes.

-- 
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