Creating migrator jobs

You start a new NTFS to Centera migration by creating a new migrator job.

To create a migrator job

  1. Run NTFSCenteraMigrator.

  2. Select option 2, Create new job.

  3. Type the number of the vault store to use as the source for the migration.

  4. Type the number of the source partition to migrate.

  5. Type the number of the Centera partition to use as the destination partition.

  6. When the utility prompts you to type the start date and end date of a range, press Enter without specifying a date, or type the year, month, and day. (Use four digits to specify the year; for example, 2006.) If you do not specify either date, the utility migrates all the savesets in the partition.

  7. When the utility prompts you for the number of worker threads to use, type a number between 1 and 25. The default is 15.

    The number of threads affects the rate at which items can be stored in the Centera. Higher numbers increase the storage rate but use more resources on the Storage Service computer.

  8. Enter the worker thread priority to use. This priority can be either of the following:

    • Below Normal. Windows gives priority to other threads, so migrator activities have lower priority than applications on the computer.

      Setting the number of worker threads to 15 or more and selecting Below Normal should give good performance when the computer is not busy with other tasks.

    • Normal. Windows gives equal priority to migrator activities and other applications.

  9. Enter the saveset sharing option to use. This option can be one of the following:

    • 0 - Use Partition property. Use the same setting as for the destination partition.

    • Force off. Saveset sharing is disabled. This increases performance at the expense of space.

    • Force on. Saveset sharing is enabled. This maximizes the storage but reduces the migration performance.

  10. When the utility prompts you for the name and location of the log file, either type the full path to the file or press Enter to use the default name and location. For example, you could try the path E:\Reports\Migration001.log. Any folder that you specify must already exist.

    By default, the NTFS to Centera Migrator creates a log file for each job in the Enterprise Vault Reports subfolder (normally C:\Program Files\Enterprise Vault\Reports). If you do not specify a log file name, the name that is used is NCM_DateAndTime.log, where DateAndTime indicates the date and time that the job was created.

  11. Choose whether to remove all references to a saveset if the saveset file no longer exists in the source partition.

  12. If a saveset has two or more sharers, choose whether to remove the unselected sharers and compact the saveset before storing it.

    If you choose not to remove unselected sharers, the utility stores the complete saveset in the Centera Clip, including multiple sharers, if present. This results in larger savesets on the Centera and hence more occupied space. The required sharer is selected when the saveset is stored and retrieved.

  13. Choose the required error handling options.

    Error wait time

    Specifies the number of seconds for which the utility waits before retrying the operation, if an error occurs. The default is 10.

    Note that the utility does not perform a retry for the following error conditions:

    • STORAGE_E_EXTRACT_CAB_HR: Error extracting Saveset file from Cab file

    • STORAGE_E_SAVESET_DECOMPRESSION: Error decompressing Saveset

    • STORAGE_E_SAVESETNOTVALID: Invalid Saveset

    For these error conditions, the utility immediately abandons processing of the saveset. However, it tries to process the saveset again when the Storage Service is restarted. (Restarting this service restarts the migration job.)

    Error count

    Specifies the maximum number of times that the utility retries processing a saveset. The default is five.

    If the utility fails to process the saveset after the maximum number of retries, it performs one of the following actions:

    • If the error appears irrecoverable, the utility abandons processing of the saveset. However, it tries to process the saveset again when the Storage Service is restarted.

    • If the error is potentially recoverable, such as a network problem, the utility pauses the thread for the error pause time (see below), and then tries to process the saveset again.

    Error pause time

    Specifies the number of minutes for which to pause the thread before trying to process a saveset again, if the utility fails to process the saveset after the maximum number of tries, but the error is potentially recoverable. The default is five.

  14. Restart the Storage Service that manages the vault store. The new job starts when the Storage Service has restarted.

More Information

Log files