Performing a Linux Workstation Capture

Use the following procedures to capture data using a Linux workstation. See Data Capture for more information on data capture options.

Prerequisites

You need to know the following information in order to capture data from a Linux workstation:

In addition, make sure that:

Installing the Capture Utility on a Linux workstation

Use the following procedure to install the Capture Utility on a Linux workstation.

  1. On a workstation, connect to the Transfer Appliance web interface.
  2. Select Linux in the Data Capture menu to download the Capture Utility installer. The Capture Utility contains configuration information specific to the Transfer Appliance from which it was downloaded. Only download the Capture Utility from the appropriate Transfer Appliance.
  3. Extract the downloaded archive.

    This creates the CaptureUtility directory containing the installer script, installer.sh.

    tar -xvzf CaptureUtility.tar.gz
    

  4. Run the installer.sh in the CaptureUtility directory:

       cd CaptureUtility
       ./installer.sh
    

    The Capture Utility tacapture.sh is installed in the same CaptureUtility directory .

Testing connectivity to the Transfer Appliance

Before running your first data capture job, confirm that the workstation can connect to the Transfer Appliance.

  1. On the Linux workstation, open a terminal window.

  2. Test connectivity with the Transfer Appliance by running tacapture.sh with the -t option.

     ./tacapture.sh -t
    
  3. You are prompted to enter the capture user password.

     Enter Transfer Appliance <Transfer Appliance IP address> "cuser" password for SSH
    

    The utility tests connectivity between the workstation and Transfer Appliance and returns results similar to the following:

    Test: Ping Transfer Appliance (123.45.67.89)...OK
    Test: SSH to Transfer Appliance (123.45.67.89:6422)...OK
    Test: Control connection with Transfer Appliance (123.45.67.89:25025)...OK
    Test: Return connection from Transfer Appliance to this workstation, (could take several minutes)...OK
    

    If the connectivity test fails, refer to the error message returned to determine the cause. The most common reason for connectivity test failure is a firewall blocking the ports needed for data capture. See Preparing the Network for details on which ports need to be open.

Performing a workstation capture from a Linux workstation

Before you start a Linux workstation capture, make sure you have the following permissions to your data source:

  • Read and execute permissions for folders.
  • Read permissions for files.

If you are only capturing data that you own, you already have the required file and folder permissions. If you want to capture data owned by other people, contact your IT administrator and ask for an account that grants you access to the data. If you are in IT administrator in charge of moving other people's data, use a backup operator or service account that has read access to all of the data.

After checking connectivity, use the Capture Utility to start a data capture job.

  1. Run tacapture.sh and specify a job name and capture target:

    ./tacapture.sh [JOB NAME] [CAPTURE DIRECTORY]
    

    where [JOB NAME] is the name of the data capture job and [CAPTURE DIRECTORY] is the directory that contains the data to capture. The Capture Utility recursively captures all data in the directories under the one specified.

    Use a meaningful job name. A job name is used to identify a capture job and the files it contains for the rest of the data migration project. A job name can contain alphanumeric, underscore, and hyphen characters.

  2. When prompted, enter the capture user password provided to you by Google:

    Enter Transfer Appliance <Transfer Appliance IP address> "cuser" password for SSH
    

    The capture job runs and displays a completion message when it is finished. Leave the terminal window open while the capture job is running, or the job will be terminated. It takes up to ten minutes for the Transfer Appliance Web interface to display a "Failed" status.

    For example, the following command creates a job named data-capture to capture the data in the directory /mnt/data and its subdirectories.

    ./tacapture.sh data-capture /mnt/data
    

Using a specified number of data capture tasks

By default, the Capture Utility uses up to 8 parallel tasks for each capture job. To use fewer parallel capture tasks, use the -m option with tacapture.sh. If you have poor network bandwidth, reduce the number of parallel tasks for each capture job.

For example, the following command creates a job named dataFactory that uses at most 6 parallel data capture tasks to capture the data in the directory /mnt/data and all subdirectories.

./tacapture.sh dataFactory /mnt/data -m 6

For more information about the options for the Capture Utility, see Capture Utility Reference.

Specifying data paths for capture

You can specify data paths for capture in one of two ways:

  • If you want to capture a single directory or file, provide a single directory when running a capture job. This method captures data from the specified directory and its subdirectories.
  • To capture multiple target files and directories, run tacapture.sh with the -f option. You must provide a line delimited text file containing absolute paths to the files you want to capture.

The following is an example of the contents of the text file you must provide when you use the -f option.

/usr/bin/app/archive/data
/etc/acme/logs/log_001
/home/Administrator/Documents

The following command creates a job dataFactory that captures data from the above targets specified in /home/user/filespec.txt.

./tacapture.sh dataFactory -f /home/user/filespec.txt

For more information about Capture Utility options, see Capture Utility Reference.

To direct the Capture Utility to skip symbolic links to files, use the -e option with tacapture.sh. For example, the following command creates a job named data-capture to capture the data in the directory /mnt/data (including its subdirectories), but skips symbolic links to files.

./tacapture.sh data-capture /mnt/data -e

For more information about the options for the Capture Utility, see Capture Utility Reference.

Capturing file metadata

The Capture Utility (tacapture.sh) preserves the following file metadata by default:

  • The user ID of the owner.
  • The group ID of the owning group.
  • The file permissions.
  • The last modification time of the file.
  • The last file access time.

For example, the following command creates a job named data-capture to capture the data and file metadata in the directory /mnt/data and all subdirectories.

C:\Program Files\TA>tacapture.sh data-capture /mnt/data1

For more information about the options for the Capture Utility, see Capture Utility Reference.

What's next

To perform other types of parallel data capture tasks, follow these instructions on a separate workstation or on the Transfer Appliance:

If your data size exceeds the capacity of a single Google Transfer Appliance, capture your data using multiple appliances in succession.

To monitor data capture jobs, see Monitoring Data Capture Jobs.

If you are done capturing data, see Preparing and Shipping Transfer Appliance.

Esta página foi útil? Conte sua opinião sobre:

Enviar comentários sobre…

Google Transfer Appliance