http://defect.opensolaris.org/bz/show_bug.cgi?id=10805
Renee Danson Sommerfeld <renee.danson at sun.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |renee.danson at sun.com
--- Comment #3 from Renee Danson Sommerfeld <renee.danson at sun.com>
2009-08-20 17:04:45 UTC ---
(In reply to comment #2)
> The problem here is that the script execution associated with enm deactivation
> (that occurs on the way to destroying the object) fails, and as a consequence
> we go to maintenance, and don't destroy the enm object. Its handle is still
> valid (since it was read prior to the destroy), and mdb shows it's still on
> the
> object list.
>
> I can reproduce this problem with an ENM with failing scripts, but notice it
> doesn't happen with an ENM that has valid scripts - if the stop script doesn't
> exist or fails we hit this error condition. The solution is not to set
> maintenance state on destroy (which we can determine from the aux state) in
> nwamd_enm_activate_deactivate_thread().
But I think we need to let the user know that a problem was encountered while
stopping the enm; putting it in maintenance state seems to be a reasonable way
of doing that. Would it make sense to block the events if it's in maintenance,
rather than just not putting it in maintenance?
--
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.
