arcserve-KB : XOsoft errors encountered trying to replicate a new local disk that was added to Windows without a system reboot.

Last Update: 2015-09-16 11:07:32 UTC


Components:XOSOFT Replication: 12.5, 15.0
Last Modified Date:    05/10/2011
Document ID:    TEC546867
Tech Document
Title:  XOsoft errors encountered trying to replicate a new local disk that was added to Windows without a system reboot.


You are trying a start a new scenario to replicate a new local disk that was added to Windows while online without rebooting. This action is supported by the OS and the hardware and you already have previously installed the CA ARCserve RHA engine and it was started prior to the addition of the volume. You would also like to run the new scenario with as little impact as possible either by avoiding a resync or downtime.


The CA ARCserve file system driver (XOMF.SYS) only queries locally attached drives at engine service startup by design so in order to replicate a new local disk the CA XOsoft engine must be restarted so that it will recognize the new volume.

There are two solutions to consider depending on your primary goal. If the new drive must be recognized immediately and even a brief period of production downtime is not permissible then Solution 1 is your only option, this will force a resync. If avoiding a resync is preferred and brief downtime is permissible Solution 2 would be your best option. Unfortunately there is no way to avoid a resync & downtime currently.

Solution 1:

  1. Stop all scenarios active on the engine being restarted.
  2. Restart the CA ARCserve RHA Engine.
  3. Restart all previously stopped scenarios.
  4. The new drive will now be recognized and you will be allowed to run a scenario containing the new disk.

Solution 2: 

This solution involves using Host Maintenance so if you are not familiar with this feature we recommend first testing Host Maintenance in a lab to familiarize yourself with this feature before using it in production:

  1. From the CA ARCserve manager toolbar go to Host Maintenance Monitor to switch to Host Maintenance view.
  2. Right-click in the blank space and choose 'Launch Host Maintenance'.
  3. Starting with the master server, from the list choose the master host of the scenarios you wish to restart the engine on and click <Next>.
  4. After a brief moment it will present you maintenance procedures, leave the default option of 'Prepare for Host Reboot'.
  5. On the right you will see a list of scenarios that will be affected by this operation, confirm all the scenarios are listed.
  6. Click <Start> then <Yes> to confirm the action and the Host Maintenance process will then begin.
  7. The wizard will then close automatically and you will then see the host you selected in the Host Maintenance Monitor window and a sub-listing of all scenarios underneath.
  8. Under the 'State' column by each scenario you will see the words 'Preparing for HM' while the engines safely flush the spool and take the shares offline to prevent outside changes during the Host Maintenance process.
  9. When the State column changes to 'Ready for HM' on ALL scenarios you are now ready to restart the master's engine.
  10. Restart the CA ARCserve RHA engine service on the master which will complete the HM process exactly as if the system was rebooted.
  11. Nothing more needs to be done on the master other than to monitor the events window on the GUI for the words 'HM process is complete. Replication is resumed without resynchronization' for confirmation that the HM process was completed successfully. That will complete the process for the master.
  12. Repeat steps 1-9 selecting the replica this time to restart the replica engine safely without requiring a resync. This step will not incur any downtime since the master is online and it will mainly be holding changes in the spool on the master until the replica engine comes back online or the service restarted.
  13. Restart the CA XOsoft engine service on the replica and once initialized it will immediately resume all scenarios without a resync and you will see the same event in the GUi as the master 'HM process is complete. Replication is resumed without resynchronization'.
  14. With both engines restarted you may now start the new scenario that replicates the new volume as the engines will now recognize the new volume that was added on the fly.

Was this article helpful?
0 out of 0 found this helpful
Have more questions? Submit a request