[kepler-dev] [Bug 4976] New: Importing run archive from a different namespace is broken

[bugzilla-daemon at ecoinformatics.org]

http://bugzilla.ecoinformatics.org/show_bug.cgi?id=4976

           Summary: Importing run archive from a different namespace is
                    broken
           Product: Kepler
           Version: 1.x dev
          Platform: Other
        OS/Version: All
            Status: ASSIGNED
          Severity: normal
          Priority: P2
         Component: workflow run manager
        AssignedTo: barseghian at nceas.ucsb.edu
        ReportedBy: barseghian at nceas.ucsb.edu
         QAContact: kepler-dev at kepler-project.org


When you import a run, the run's execId is set to the next available execId,
and we attempt to increment the run's revision. Subsequent tag events on this
run also increment the revision. However, if this run comes from a different
namespace, it gets an entirely new lsid. There are a few problems with this at
the moment:
- a subsequent export of this imported run will not contain the derivedFrom
lsid. This is not currently stored in provenance. We can't tell who originally
created the run.
- opening the kar again during a subsequent kepler session will again import
the run with a different lsid even though it's already been imported, because
the lookup to avoid this is based on the original lsid. So you end up with two
runs that are the same in all ways but run lsid. This bug doesn't happen during
the same kepler session.

I think the solution to these issues is to make sure WorkflowRuns are getting
and storing referral list, and to then store and utilize the referral list in
provenance.

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