13 Sep, 2007

6 commits


12 Sep, 2007

6 commits

  • 1335: define, document, review CLI to register a server in a synch
    1336: CLI implementation to register a server in a synchronization
    1878: provide a CLI to initialize a replication topology from a gi
    2201: ads-trustore missing when not configuring replication with t
    2250: Setup and dsreplication utilities should configure schema sy
    
    
    Extra: do not show dc=replicationchanges suffix in the replication panel of the setup, nor propose it to replicate.
    fix some bugs related to duplicate error messages displayed in dsreplication interactive mode.
    
    
    
    git-svn-id: https://svn.forgerock.org/opendj/trunk@3031 41b1ffd8-f28e-4786-ab96-9950f0a78031
    jvergara
     
  • git-svn-id: https://svn.forgerock.org/opendj/trunk@3030 41b1ffd8-f28e-4786-ab96-9950f0a78031
    coulbeck
     
  • …sconfig would just bomb out. With this change dsconfig will either give the user the opportunity to re-edit and fix any mis-configured properties of the component (for create-xxx and set-xxx-prop interactive modes), or drop them back to the component menu (for delete-xxx interactive mode).
    
    dsconfig will still bomb out when it encounters a communication or security related problem.
    
    
    git-svn-id: https://svn.forgerock.org/opendj/trunk@3029 41b1ffd8-f28e-4786-ab96-9950f0a78031
    matthew_swift
     
  • …ere an applyConfigurationDelete was being invoked and throwing an IllegalStateException for changes that were not applicable to it.
    
    
    git-svn-id: https://svn.forgerock.org/opendj/trunk@3028 41b1ffd8-f28e-4786-ab96-9950f0a78031
    matthew_swift
     
  • git-svn-id: https://svn.forgerock.org/opendj/trunk@3025 41b1ffd8-f28e-4786-ab96-9950f0a78031
    pgamba
     
  • - Update the server code to replace simple uses of ReentrantLock involving
      {lock, doSomething, unlock} in the same method to use the synchronized
      keyword instead of a ReentrantLock object.  Using the synchronized keyword is
      actually a little faster, and also simplifies the code since it's no longer
      necessary to use a finally block to ensure that the lock is released.
    
    - Update the AsynchronousTextWriter to use the LinkedBlockingQueue.drainTo()
      method to attempt to get multiple messages at once, rather than always using
      poll() to get one message at a time.
    
    
    git-svn-id: https://svn.forgerock.org/opendj/trunk@3024 41b1ffd8-f28e-4786-ab96-9950f0a78031
    neil_a_wilson
     

11 Sep, 2007

9 commits

  • Previous to this change referential integrity was only enforced when a referencing (aggregating) component was added or modified. The support did not prevent a referenced component from being disabled or deleted. This change adds this remaining support:
    
    * a component cannot be deleted if it is referenced by one or more components
    
    * a component cannot be disabled if it is referenced by one or more components (it is possible to restrict this constraint so that it only applies when the referencing component(s) are enabled)
    
    This implementation only enforces referential integrity for referencing components which have listeners associated with them. For example, if component A references component B, then referential integrity will only be enforced automatically if component A has a change listener registered against it, or if it is "added" using an add listener that was registered against its parent. In effect, referential integrity is only enforced for components which are in use or are about to be used.
    
    
    git-svn-id: https://svn.forgerock.org/opendj/trunk@3023 41b1ffd8-f28e-4786-ab96-9950f0a78031
    matthew_swift
     
  • git-svn-id: https://svn.forgerock.org/opendj/trunk@3021 41b1ffd8-f28e-4786-ab96-9950f0a78031
    matthew_swift
     
  • …op background thread if needed.
    
    git-svn-id: https://svn.forgerock.org/opendj/trunk@3020 41b1ffd8-f28e-4786-ab96-9950f0a78031
    dugan
     
  • 
    git-svn-id: https://svn.forgerock.org/opendj/trunk@3018 41b1ffd8-f28e-4786-ab96-9950f0a78031
    neil_a_wilson
     
  • git-svn-id: https://svn.forgerock.org/opendj/trunk@3017 41b1ffd8-f28e-4786-ab96-9950f0a78031
    lutoff
     
  • git-svn-id: https://svn.forgerock.org/opendj/trunk@3013 41b1ffd8-f28e-4786-ab96-9950f0a78031
    matthew_swift
     
  • may not be properly notified when the associated entry is updated.
    
    
    git-svn-id: https://svn.forgerock.org/opendj/trunk@3008 41b1ffd8-f28e-4786-ab96-9950f0a78031
    neil_a_wilson
     
  • FIFO ordering.
    
    
    git-svn-id: https://svn.forgerock.org/opendj/trunk@3007 41b1ffd8-f28e-4786-ab96-9950f0a78031
    neil_a_wilson
     
  • improvements, including:
    
    - Replace occurrences of StringBuffer with StringBuilder.  As of Java 5,
      StringBuilder is almost always a better choice than StringBuffer because it
      provides all the same methods but doesn't have the synchronization overhead.
    
    - Update the JE backend DN2URI class (which is responsible for keeping track of
      smart referrals in the database) to keep track of whether there are any
      referrals in the underlying database so that it doesn't go to the DB if it
      can be certain that there aren't any referrals to get.
    
    - Update the JE backend ImportJob class to provide a simple cache for parent ID
      lookups so that it can avoid the need to go to the underlying DB to get the
      entry ID for a parent entry if it's contained in the cache.
    
    - Update the salted password storage scheme variants so that they use an
      instance of java.util.Random rather than java.security.SecureRandom to
      generate the salt.  There really isn't a need for a cryptographic-quality
      random number generator for salt generation, and the java.util.Random
      implementation is much faster than the java.security.SecureRandom
      implementation.
    
    - Update the DN class to return a reference to the precomputed NULL_DN instead
      of creating a new DN with zero RDN components when decoding DNs from strings
      or octet strings.  Also, use LinkedList rather than ArrayList for temporary
      storage of RDN values to avoid the hit of allocating memory that we may not
      need.
    
    
    - Update the Entry class to cache the result of LDIFImportConfig.typesOnly(),
      which has been observed to slightly improve MakeLDIF performance.
    
    - If Java 6 is available, then prefer using it to set file permissions over
      Runtime.exec() on UNIX systems.  Even though it's potentially less
      fine-grained than using exec to call chmod, it's faster and safer to use the
      Java methods if they're available.
    
    
    git-svn-id: https://svn.forgerock.org/opendj/trunk@3006 41b1ffd8-f28e-4786-ab96-9950f0a78031
    neil_a_wilson
     

10 Sep, 2007

4 commits


09 Sep, 2007

3 commits

  • will be invoked.  There are two basic changes:
    
    - Add a new ds-cfg-invoke-for-internal-operations configuration attribute for
      all plugins, which indicates whether the plugin should be invoked for
      internal operations.  If this is false, then the plugin will only be invoked
      for externally-requested operations.
    
    - Add four new plugin types:  postSynchronizationAdd,
      postSynchronizationDelete, postSynchronizationModify, and
      postSynchronizationModifyDN.  These allow a plugin to perform a limited set
      of processing for changes that are successfully applied through
      synchronization.
    
    
    The unique attribute plugin has also been updated to support the
    post-synchronization plugin types so that if a conflict is introduced
    concurrently on two different servers within the propagation delay, an
    administrative alert will be generated to indicate that manual intervention is
    required to address the problem.
    
    Finally, ensure that audit logging is enabled during the unit tests, and
    update the audit logger to include the connection ID and operation ID for the
    operation being logged.
    
    
    OpenDS Issue Number:  2057
    
    
    git-svn-id: https://svn.forgerock.org/opendj/trunk@2999 41b1ffd8-f28e-4786-ab96-9950f0a78031
    neil_a_wilson
     
  • implement generic entry cache monitor provider which allows any entry cache implementation current or future
      to provide a common or custom set of entry cache state data. update all existing entry cache implementations
      so that they provide their common state data to the entry cache monitor provider.
    
    
    git-svn-id: https://svn.forgerock.org/opendj/trunk@2998 41b1ffd8-f28e-4786-ab96-9950f0a78031
    abobrov
     
  • CryptoManager
    Refactor to separate key and key entry generation from key retrieval.
    
    
    git-svn-id: https://svn.forgerock.org/opendj/trunk@2997 41b1ffd8-f28e-4786-ab96-9950f0a78031
    david_page
     

08 Sep, 2007

1 commit

  • CryptoManager encryption
    
    Allow key size different from block size for block ciphers that support it (e.g., Blowfish as in Blowfish/CFB/NoPadding", 128) and block ciphers modes that use no initialization vector (e.g., ECB as in "DES/ECB/PKCS5Padding", 64). Note that this is accomplished without adding an initialization vector parameter to the encryption API. Instead, if the Cipher.getBlockSize is non-zero, the Cipher object is first instantiated with that size initialization vector, and if that fails, initialized without an initialization vector.
    
    
    git-svn-id: https://svn.forgerock.org/opendj/trunk@2996 41b1ffd8-f28e-4786-ab96-9950f0a78031
    david_page
     

07 Sep, 2007

6 commits


06 Sep, 2007

5 commits