http://defect.opensolaris.org/bz/show_bug.cgi?id=10277
amaguire <alan.maguire at sun.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
AssignedTo|nwam-dev at opensolaris.org |alan.maguire at sun.com
Status|INCOMPLETE |CAUSEKNOWN
--- Comment #4 from amaguire <alan.maguire at sun.com> 2009-07-27 09:02:18 ---
(In reply to comment #3)
> I'm seeing a failure in v4 configuration with the latest bits too, but it
> seems
> to be a side-effect of the ioctl hang (leading to deadlock, described in
> 10091,
> which I've reopened) during IPv6 plumb. The static address will never be
> assigned since the IP NCU is locked during the hanging ioctl operation, and as
> a consequence we can't reacquire the lock to set the IP address. If a v4-only
> NCU is configured, static addresses are correctly assigned, so I think this
> bug
> can be closed.
For some reason, the deadlock problem has gone away for me and now I'm seeing
the failure to assign static v4 addresses. The reason looks to be that if both
v4 and v6 are to be configured, we
- free any addresses stored in the interface ncu structure
- add the list of v4 addresses to be configured
- free any addresses stored in the interface ncu structure
- add the list of v6 addresses to be configured
So obviously we need to move the freeing of existing addresses
stored from the v4/v6 logic in populate_ip_ncu_properties(),
since otherwise we simply throw away any static v4 addresses
we've stored when we go to store v6 static addresses.
--
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.
