1. 18 Jun, 2013 1 commit
  2. 24 May, 2013 1 commit
  3. 23 May, 2013 1 commit
    • Steven Murray's avatar
      bug #101536: The tapebridged daemon does not handle zero length files for migration correctly. · 2804d8d4
      Steven Murray authored
      The new unit tests (there are now only 3) for the PendingMigrationsStore class
      have been improved and the PendingMigrationsStore class itself has been
      sucessfully modified to pass the tests.
      
      The next step is propagate the new SessionException up through the call stack
      until the BridgeProtocolEngine::runRtcpdSession() method and modify the method
      so that it extracts the SessionError wrapped by the SessionException and
      appends it to BridgeProtocolEngine::m_sessionErrors memebr variable.
      2804d8d4
  4. 22 May, 2013 1 commit
  5. 03 May, 2013 1 commit
  6. 30 Apr, 2013 1 commit
  7. 28 Mar, 2013 1 commit
  8. 21 Mar, 2013 1 commit
  9. 03 Dec, 2012 3 commits
  10. 03 Oct, 2012 1 commit
  11. 02 Oct, 2012 3 commits
    • Steven Murray's avatar
      bug #88114: RFE: wrong compression for LTO5 and Oracle T10KC drives · a6970f1d
      Steven Murray authored
      Even though I unit tested the following method I never actually called it:
      
      BridgeProtocolEngine::correctMigrationCompressionStatistics()
      
      The method is now called at the appropriate moment and the writetp
      command line tool now displays the compressed size of the files it
      migrates to tape (thanks to Victor for this modification), e.g.:
      
      Oct 02 12:34:58 Migrated "lxc2dev5d1.cern.ch:/var/tmp/copyTape/TestZero" size=10485760 checksum=0x9600001 compressed size=46080
      a6970f1d
    • Steven Murray's avatar
      bug #88114: RFE: wrong compression for LTO5 and Oracle T10KC drives · af919cff
      Steven Murray authored
      I have improved the tapebridged side of the batch compresion logic by adding
      code to deal with the unlikely case that the sum of the original file sizes of
      a batch of migrated files is zero bytes.  This should never happen, however the
      code should not crash if presented with such a case.
      
      The following two new unit tests explain the expected behaviour and pass with
      success.
      
        CPPUNIT_TEST(testCalcMigrationCompressionRatioWithZeroFileSize);
        CPPUNIT_TEST(testCorrectMigrationCompressionStatisticsWithZeroFileSize);
      af919cff
    • Steven Murray's avatar
      bug #88114: RFE: wrong compression for LTO5 and Oracle T10KC drives · 2d34b669
      Steven Murray authored
      Implemented the tapebridged daemon side of the bulk compression logic.  This
      implementation includes the following 2 unit tests that pass successfully:
      
        CPPUNIT_TEST(testCalcMigrationCompressionRatio);
        CPPUNIT_TEST(testCorrectMigrationCompressionStatistics);
      2d34b669
  12. 05 Sep, 2012 1 commit
  13. 03 Sep, 2012 1 commit
  14. 15 Aug, 2012 1 commit
  15. 14 Aug, 2012 1 commit
  16. 07 Aug, 2012 2 commits
    • Steven Murray's avatar
      bug #96536: rmlabelinfo() of tape/initlabel.c corrupts memory if called more than once · 8cb81c63
      Steven Murray authored
      The BridgeProtocolEngine::rtcpTapeErrReqRtcpdCallback() method currently
      aborts the tape session if it detects an error by throwing an exception.
      This is too abrupt and should be changed to trying to gracefully shutdown
      the rtcpd session.
      
      This commit introdcues the graceful shutdown of the rtcpd session in the
      event of an error detected by the
      BridgeProtocolEngine::rtcpTapeErrReqRtcpdCallback() method.
      8cb81c63
    • Steven Murray's avatar
      bug #96536: rmlabelinfo() of tape/initlabel.c corrupts memory if called more than once · b8398ac0
      Steven Murray authored
      The castor::tape::tapebridge::BridgeProtocolEngine::rtcpTapeErrReqRtcpdCallback()
      method currently hinders the shutdown of rtcpd in error conditions by
      acknowledging an rtcpd RTCP_TAPEERR_REQ with the acknowledge of an
      RTCP_TAPE_REQ message.  Rtcpd detects a protocol error which confuses the
      shutdown sequence.
      
      This commit correctly the acknowledge message send by the tapebridged daemon.
      b8398ac0
  17. 25 Jul, 2012 1 commit
  18. 11 Jul, 2012 2 commits
  19. 06 Jun, 2012 1 commit
  20. 02 May, 2012 1 commit
  21. 24 Apr, 2012 1 commit
    • Steven Murray's avatar
      bug #92460: tapebridged should gracefully shutdown a migration tape-session... · 78d241db
      Steven Murray authored
      bug #92460: tapebridged should gracefully shutdown a migration tape-session when tapegatewayd reports a disabled tape
      
      Refactored the tapebridged daemon and created the following unit-test that
      recreates this bug:
      
      castor::tape::tapebridge::BridgeProtocolEngineTest::testMigrationToDisabledTapeUsingLocalDomain
      
      The refactoring was necessary in order to create the unit-test.  During the
      refactoring I modified the logging of the tapebridged daemon so that the very
      first session error detected is always logged immediately.  This will help
      debug problems in the future if need be.
      
      78d241db
  22. 12 Apr, 2012 2 commits
  23. 17 Feb, 2012 1 commit
    • Steven Murray's avatar
      bug #90313: RFE: tapebridged should request more files to transfer in bulk · 62f42172
      Steven Murray authored
      Fixed a stack-overflow bug in the CASTOR 2.1.12-2 internal release of the
      tapebridged daemon.  The bug was caused by an unintentionally recursive
      function.  The bug was not detected on my development system but was detected
      on castorcert5 because the bug only manifested itself when an error was
      reported to the tapebridged daemon by the rtcpd daemon.
      62f42172
  24. 10 Feb, 2012 1 commit
  25. 11 Jan, 2012 1 commit
  26. 10 Jan, 2012 1 commit
    • Steven Murray's avatar
      bug #90313: RFE: tapebridged should request more files to transfer in bulk · 8aa7b55d
      Steven Murray authored
      The implementation of this RFE is going to take a relatively long period of
      time in that there will be several commits to SVN before it is completed.
      
      This commit adds some helper classes to the tapebridged source-code base.
      These classes are tested by the unit-tests of the test/unittest directory but
      are not actively used by the tapebridged daemon and so will not effect any
      release that may need to be made of the trunk.
      8aa7b55d
  27. 02 Dec, 2011 4 commits
  28. 28 Nov, 2011 2 commits
  29. 25 Nov, 2011 1 commit