[kepler-dev] [Bug 5420] Two types of NPE issues when 2.2.0 installed and run by a non-admin windows user

[bugzilla-daemon at ecoinformatics.org]

http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5420

Derik Barseghian <barseghian at nceas.ucsb.edu> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |WONTFIX

--- Comment #1 from Derik Barseghian <barseghian at nceas.ucsb.edu> 2011-06-13 12:57:30 PDT ---
David looked into this bug. I've pasted his assessment is below, essentially I
managed to stumble on a  bug that will be unlikely for normal users. This
morning I did a few tests to verify his assessment. Since this bug is very
unlikely to occur, I am marking it WONTFIX. 
I think it really comes down our need to improve our configuration system, and
I think bug#5129 has become the defacto bug for that.
----
This whole set of conditions is rather unlikely to come up for "normal users."
For a normal user to replicate this error, they would have to do the following:

(1.) Install Kepler 2.2.
(2.) Run Kepler 2.2, but avoid using either the "Search Components" box to
search for a workflow component or the "Sources" button.
(3.) Switch to Kepler 2.0, and use the "Search Components" box.
(4.) Switch back to Kepler 2.2 and use the "Search Components" box and the
"Sources" button to get both errors you reported.

As you can see, this has nothing to do with Windows, permissions, or any of the
rest of it. Further, it is going to be very rare for ordinary users to
experience this. For ordinary users to experience this they would have to be of
the sort that switched back and forth between Kepler 2.0 and Kepler 2.2 AFTER
not only installing but also running Kepler 2.2 AND they would have to never
have used the Search Components box in Kepler 2.2, meaning that they probably
never constructed a workflow of any sort in Kepler 2.2, BUT they do use this
same functionality in Kepler 2.0. It sounds rather unlikely to me. I would not
be surprised if this bug was not replicated by a single user.

Now, it is possible to eliminate even the very rare possibility of this bug
happening to ordinary users entirely by changing the design of the
configuration system so that each release of Kepler has its own configuration
area. But we have discussed this design before and decided against it.

I do not believe that this bug that is rarely likely to come up in practice
(and which may not be able to be addressed without a significant design change
to the configuration system) is a release blocker.
---

-- 
Configure bugmail: http://bugzilla.ecoinformatics.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.