Derby
  1. Derby
  2. DERBY-5940

Disable securityMechanism=8 in NSSecurityMechanismTest on ARM platforms

    Details

    • Type: Task Task
    • Status: Closed
    • Priority: Major Major
    • Resolution: Fixed
    • Affects Version/s: 10.10.1.1
    • Fix Version/s: 10.10.1.1
    • Component/s: Test
    • Labels:
      None
    • Environment:
      ARM platforms

      Description

      The security mechanism strong password substitution (securityMechanism=8) uses code that relies on entropy, which seems to be in short supply on some ARM systems. This includes the ARM server farm available in the ASF build/test farm.

      I will disable tests using strong password substitution in NSSecurityMechanismTest on ARM platforms to be able to run the test to completion within a reasonable time.
      DERBY-3106 and DERBY-5651 are relevant.

        Activity

        Hide
        Kristian Waagan added a comment -

        suites.All on ARM platform completed.
        Closing issue.

        Show
        Kristian Waagan added a comment - suites.All on ARM platform completed. Closing issue.
        Hide
        Kristian Waagan added a comment -

        Thanks, Knut.

        I went ahead and committed patch 1a to trunk with revision 1392819.
        I've started run #15 [1] of suites.All on ARM (it takes 7+ hours to complete).

        [1] https://builds.apache.org/job/Derby-trunk-suites.All-ARM/

        Show
        Kristian Waagan added a comment - Thanks, Knut. I went ahead and committed patch 1a to trunk with revision 1392819. I've started run #15 [1] of suites.All on ARM (it takes 7+ hours to complete). [1] https://builds.apache.org/job/Derby-trunk-suites.All-ARM/
        Hide
        Knut Anders Hatlen added a comment -

        I've seen this problem too. As I mentioned in my comment on DERBY-3106, I don't think it's worthwhile fixing the underlying problem, since we no longer recommend the use of securityMechanism=8. So disabling the test cases on ARM sounds fine to me. +1 to the approach in the 1a patch.

        Show
        Knut Anders Hatlen added a comment - I've seen this problem too. As I mentioned in my comment on DERBY-3106 , I don't think it's worthwhile fixing the underlying problem, since we no longer recommend the use of securityMechanism=8. So disabling the test cases on ARM sounds fine to me. +1 to the approach in the 1a patch.
        Hide
        Kristian Waagan added a comment -

        I hope to commit this patch shortly.

        Show
        Kristian Waagan added a comment - I hope to commit this patch shortly.
        Hide
        Kristian Waagan added a comment -

        Attaching patch 1a, which disables securityMechanism=8 in NSSecurityMechanismTest.
        I verified the patch by tracing NetConnection.flowConnect. Forcing hasSufficientEntropy to false, the number of flows went down from 231 to 132.
        I've also tested a very similar patch on the ASF ARM build farm.

        Patch ready for review.

        Show
        Kristian Waagan added a comment - Attaching patch 1a, which disables securityMechanism=8 in NSSecurityMechanismTest. I verified the patch by tracing NetConnection.flowConnect. Forcing hasSufficientEntropy to false, the number of flows went down from 231 to 132. I've also tested a very similar patch on the ASF ARM build farm. Patch ready for review.

          People

          • Assignee:
            Kristian Waagan
            Reporter:
            Kristian Waagan
          • Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

            • Created:
              Updated:
              Resolved:

              Development