Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-8159

Overflow converting to int error in metadata.add

    XMLWordPrintable

    Details

    • Type: Bug
    • Status: Done
    • Resolution: Done
    • Fix Version/s: None
    • Component/s: pipe_base
    • Labels:
      None
    • Team:
      Architecture

      Description

      When running multibandDriver.py on a fairly large dataset (see DM-6816), I am encountering many error messages along the lines of:

      5547 FATAL multiBandDriver.measureCoaddSources.deblend: SourceDeblendTask.metadata.add(name='runEndInvoluntaryContextSwitches', value=13810175634) failed with error=Overflow error converting to int
      7769 FATAL multiBandDriver.measureCoaddSources.match.refObjLoader: LoadAstrometryNetObjectsTask.metadata.add(name='loadSkyCircleStartInvoluntaryContextSwitches', value=9892578161) failed with error=Overflow error converting to int
      

      So it seems this is a bug in daf_base's PropertySet/PropertyList add function. Might this be related to the py3 conversion?

      Also note that this is not FATAL in the sense that it does not bring the whole run to a halt.

        Attachments

          Activity

          Hide
          frossie Frossie Economou added a comment -

          Thank you KT!

          Show
          frossie Frossie Economou added a comment - Thank you KT!
          Hide
          ktl Kian-Tat Lim added a comment -

          Jenkins failed on Centos7 Py2, but I think it was a transient error in skymap as Flake8 checking was being turned on.

          Show
          ktl Kian-Tat Lim added a comment - Jenkins failed on Centos7 Py2, but I think it was a transient error in skymap as Flake8 checking was being turned on.
          Hide
          price Paul Price added a comment -

          The original obj.metadata.add is now unprotected by the try block. Do you believe all reasons for embedding it in a try block in the first place have been dealt with by this change?

          Show
          price Paul Price added a comment - The original obj.metadata.add is now unprotected by the try block. Do you believe all reasons for embedding it in a try block in the first place have been dealt with by this change?
          Hide
          ktl Kian-Tat Lim added a comment -

          Yes, based on reading the original code and its motivation, and I believe that any other reasons indicate severe enough problems (hinted at by the previous FATAL log message) that should indeed throw exceptions rather than having them be trapped.

          Show
          ktl Kian-Tat Lim added a comment - Yes, based on reading the original code and its motivation, and I believe that any other reasons indicate severe enough problems (hinted at by the previous FATAL log message) that should indeed throw exceptions rather than having them be trapped.
          Hide
          ktl Kian-Tat Lim added a comment -

          Merged in a695220.

          Show
          ktl Kian-Tat Lim added a comment - Merged in a695220.

            People

            Assignee:
            ktl Kian-Tat Lim
            Reporter:
            lauren Lauren MacArthur
            Reviewers:
            Paul Price
            Watchers:
            Fred Moolekamp, Frossie Economou, Jim Bosch, John Swinbank, Kian-Tat Lim, Lauren MacArthur, Paul Price, Pim Schellart [X] (Inactive), Tim Jenness, Yusra AlSayyad
            Votes:
            0 Vote for this issue
            Watchers:
            10 Start watching this issue

              Dates

              Created:
              Updated:
              Resolved:

                Jenkins

                No builds found.