PlateSpin Migrate 9.0.2 Release Notes

September 30, 2010

1.0 About this Release

Version 9.0.2 is the follow-on release after the PlateSpin Migrate 9 (English only) and 9.0.1 (international) releases. It introduces a number of new features, enhancements, and bug fixes.

For Release Notes documents that accompanied previous 9.x.x releases, visit the PlateSpin Migrate 9 Documentation Web Site and check Previous Releases at the bottom of the main page.

2.0 New Features in Version 9.0.2

  • Migration Support for Additional Source Workload Types: This release extends migration support to the following operating systems:

    • Microsoft Windows Server 2008 R2

    • Microsoft Windows 7

  • Migration Support for Additional VMware Targets: This release extends migration support to the following VMware targets:

    • VMware ESX Server 4.1

    • VMware vCenter 4.1

  • Capability to Discover and Access VM platforms by Using Windows Domain Credentials: You can now discover and access the following VM targets by using your Windows domain credentials:

    • VMware ESX 4.0, 4.1

    • VMware vCenter 4.0, 4.1

3.0 Enhancements in Version 9.0.2

  • Support for Bonded Networks: This release supports Linux source workloads with bonded NICs (link aggregation).

4.0 Bugs Fixed in Version 9.0.2

  • 609918 Problem with automatic VMware tools installation during P2V: During a P2V migration of a SUSE Linux Enterprise Server (SLES) 11 SP1 workload, the virtual CD-ROM device might fail to mount on the recovery workload, impacting the installation of the VMware Tools software and causing warnings. This issue has been fixed.

  • 609885 (ESX 4) Non-ASCII characters in the Migrate Server hostname preventing workload migrations: Migrations to a VMware ESX 4 system failed if your PlateSpin Migrate Server had non-ASCII characters in its hostname. This issue has been fixed.

  • 632022 Chinese characters in Server hostname causing migration problems: Chinese characters in a PlateSpin Migrate Server hostname resulted in migration job failure at the Setup Controller on Source Server step. This issue has been fixed.

  • 634901 Target dynamic disks without drive letters causing Server Sync failure: Server Sync jobs failed with the error Disks with target serial numbers and PlateSpinVolumeMap.xml not found error when the target had dynamic disks without drive letters. This issue has been fixed.

  • 613661 Problem with semi-automated X2P migrations of Windows Server 2008: When attempting to migrate a Windows Server 2008 workload to a VM with two disks registered as a physical target, selecting the second disk as the target datastore resulted in a job failure with a 'bcdedit.exe' returned a failure exit code error at the PrepareOsToBoot step. This issue has been fixed.

  • 639829 Server with multiple IP addresses failing to register a Linux X2P target: Registration of a physical target for a Linux X2P migration did not work properly if the PlateSpin Migrate Server was configured with multiple IP addresses. This issue has been fixed.

  • 637066 Invalid characters in a workload’s machine.xml file causing discovery disruption: Invalid XML characters, occasionally found among source workloads’ service descriptions, were improperly filtered out of a workload’s machine.xml file, resulting in discovery or refresh problems. This issue has been fixed.

  • 634854 Outdated target VM network information failing Server Sync jobs: The validation mechanism failed to capture VM network changes since the VM’s addition as a Server Sync target. This issue has been fixed.

  • 626422 Problem Discovering ESX Servers through vCenter with dvSwitch: When running a discovery job against a VMware vCenter that had a vNetwork Distributed Switch (dvSwitch, vDS) configured on it, the only ESX Servers in the cluster that were being actually discovered were the first one along with those that were in maintenance mode, resulting in warnings. This issue was fixed in 9.0.1 and migrated to 9.0.2.

5.0 Known Issues in Version 9.0.2

  • No software RAID support for Linux workloads: PlateSpin Migrate does not support Linux workloads with volumes on software RAID.

  • 493589 (Windows sources) Non-default per-volume VSS settings are not preserved after migration: This issue is under consideration for an upcoming fix.

  • 548208 Discovery failure if vCenter has an ESX server with no datastore: VMware vCenter discovery fails if an individual ESX server managed by vCenter has no datastore configured.

    Workaround: Add a datastore to the ESX server and try again.

  • 505426 (ESX4) No warning or error on wrong vCPU selection: If the number of the requested vCPUs exceeds the number of physical CPUs on the ESX 4 host, the requested number is ignored and the target VM is created with a single vCPU without a warning. This issue is under consideration for an upcoming fix.

  • 604320 (Linux to ESX 4) Problem completing migration if the source OS has autologin or CD automount features enabled: The migration is also affected if you log in to the target during the job’s Configuration step.

    Workaround: Disable the autologin and CD automount features on the source; avoid logging in to the target workload prior to the completion of the migration.

  • 619942 Failure to execute a post-migration script with Unicode characters in the filename: If you use Unicode characters in the filename of your post-migration script, the script fails to execute.

    Workaround: Use only ASCII characters when naming a post-migration action.

  • 597846 (NT 4 to ESX 4) No post-migration networking: Upon migrating a Windows NT 4 workload to a VMware ESX 4 system, the target VM might have no networking.

    Workaround:

    1. Remove any Ethernet adapters and restart the workload.

    2. Add an Ethernet adapter by using the Have Disk option and pointing the system to the vmxnet folder of VMware Tools installation directory (the default path is C:\Program Files\VMWARE\VMware Tools\DRIVERS\vmxnet). Select the VMware Virtual Ethernet Adapter option and ignore any warnings.

    3. Configure the Ethernet adapter as required and restart the workload.