http://defect.opensolaris.org/bz/show_bug.cgi?id=11437
Summary: after booting, both wired and wireless have IP plumbed
for the Automatic NCP
Classification: Development
Product: nwam
Version: unspecified
Platform: ANY/Generic
OS/Version: All
Status: ACCEPTED
Severity: minor
Priority: P1
Component: ON daemon
AssignedTo: alan.maguire at sun.com
ReportedBy: alan.maguire at sun.com
QAContact: nwam-dev at opensolaris.org
--- Comment #0 from amaguire <alan.maguire at sun.com> 2009-09-18 16:11:57 UTC
---
I'm seeing this happen on boot only. What seems to happen is on boot we get an
initial link state up event for e1000g0, which causes us to propogate an up
link state to the IP NCU. Then we get the link down, which causes us to
propogate the link down to e1000g0, but right after we switch to
offline*,waiting for addr, so it seems like somehow we're losing the
online*,conditions unmet state change and still assuming we're supposed to
configure e1000g0.
The problem appears to be in nwamd_interface_ncu_up(), which if the ncu has no
configured addresses, pushes the state into online*, initialized. I believe
this should be online*, conditions unmet. online*/initialized is an unexpected
state/aux state combination, and will lead us to re-plumb and go to
offline*,waiting for addr.
I think the right answer here may be to rename nwamd_interface_ncu_up() to
nwamd_interface_ncu_up_down(), and use it to bring up the NCU if up is true, as
it is in all cases expect where we call it from an RTM_DELADDR when handing
that IF_STATE event. In that case, up is false and we should bring the ncu to
online*,conditions unmet.
This fix _may_ address 11092 also - need to test it more first.
--
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.
