Waiting for Controller to Start - Controller Connection Not Established/Broken

  • 7920405
  • 27-Oct-2005
  • 27-Apr-2012

Environment

PlateSpin Migrate, Protect, or Forge

Situation

This article outlines information on troubleshooting the following error(s) that may appear during the Take Control of source or target virtual/physical machine steps of a conversion:

Waiting for Controller to start(Failed) Controller Connection Not Established

Controller Connection Broken

Unable to start the Heartbeat Service

Resolution

The occurrence of this error generally indicates that either: 
  1. The network settings for the temporary IP addresses under the Job Configuration -> Take Control section were not properly configured   
  2. There was a possible network outage that prevented the source/target machine from communicating with the PowerConvert Server 
  3. The source/target machine was not able to fully boot up under PlateSpin control

The following is a checklist of items to verify when troubleshooting this particular error:

  • What state is the source server or target VM/physical server currently in?  Is it under PlateSpin control? (i.e.  Is it sitting at a DOS prompt?) 
  • If performing a Live Transfer conversion of the source machine and the job fails during "Verifying Controller Connection", please see TID 7920835
  • If the source/target machine is not at a DOS prompt, note the state of the machine and contact PlateSpin support 
  • If the source machine is displaying a STOP 0x7B Inaccessible Boot Device error or similar this most likely indicates that the PlateSpin take control image does not have proper driver support for your SCSI/RAID controller.  To workaround this issue, the user can try rebooting the source machine and press F6 when prompted to manually add the 3rd party drivers provided by their hardware vendor.  Please note that the PlateSpin take control image requires Windows 2003 based drivers.
  • Check to see if the Z:\controller\eventlog\active.log file exists.  If so, verify whether or not the log contains any errors such as "Unable to start the Heartbeat Service".
  • Check to see if the RegisterUnderControlMachine.log file exists.  This file will either be located in Z:\ (if the machine is a target VM or target physical machine) or X:\I386\system32\platespin\registration (if the machine is a source machine).  If the file exists, verify whether or not the log contains any errors.
  • Type "ipconfig" at the DOS prompt to see if the source server/target VM has a valid IP address and not a 169.xxx.xxx.xxx autoconfiguration address.  You will also want to confirm if the IP address assigned to the source server/target VM corresponds with the network settings that was initially configured in the Job Configuration screen of PowerConvert.

    If the source server or target VM do not have a valid IP address, then the following options are available:

    1)  Restart the Conversion job with a valid IP address configured
    2)  Use a 3rd party freeware utility such as
    PE Network Configurator which can be copied and executed from the floppy drive to 
         manually set the IP address

  • If the source server and target VM/physical server each have valid IP addresses, check to see if the source and target VM/physical server can ping each other successfully.  Also, verify that both can also ping the IP address of the PowerConvert Server and vice-versa.
  • If the source server and target VM/physical server have valid IP addresses and can ping the IP address of the PowerConvert Server, but the error "Underlying connection was closed" is displayed, try typing "registermachine" at the command prompt to re-initiate the download of the PlateSpin Controller
  • If the source/target machines have more than one network adapter, ensure that the IP address specified in the Job Configuration is assigned to the correct network adapter or virtual network adapter (especially important if there is more than one virtual network configured on the target virtual host)
  • There should be a PlateSpin service called OfxController running on both the source server and target VM.  Verify that this service is running by typing the following:  1) net start ofxcontroller (on Windows source servers or VMware GSX/Microsoft Virtual Server platforms)   OR   2)  service ofxcontrollerd status (for VMware ESX Server)
  • Ensure that ASP.NET, IIS Admin Service and World Wide Web Publishing Service are all running on the PowerConvert server.  You can verify this by accessing the Event Viewer and Service Manager.


**If the problems persist after attempting a conversion in troubleshooting mode, please contact PlateSpin support at support@platespin.com and include the following with your email:

  • The output from the "ipconfig /all" command.  The output can be redirected to a .txt file by typing "ipconfig /all > ipconfig.txt"
  • The Z:\controller\eventlog\active.log and X:\I386\system32\platespin\registration\RegisterUnderControlMachine.log or Z:\RegisterUnderControlMachine.log file from the source or target machine (respectively) that is experiencing the error.  To acquire these files copy them to a floppy drive while the machine is under PlateSpin control.
  • The details of any error messages displayed (if the machine was unable to successfully bootup)
  • The PowerConvert Server database files.  For instructions on sending the database files, please see the Knowledge Base article titled TID 7920288

If you have questions about this article please contact PlateSpin Support at support.powerconvert@platespin.com