Troubleshooting virtualization issues

Ethernet adapter not deleted from LPAR during rollback

In an IBM Frame, a rollback does not delete an Ethernet adapter from the logical partition (LPAR), even though the results report a successful (passed) rollback. This happens if the new Ethernet adapter is created with a slot ID available at creation time rather than the one named in the BLpackage.

This issue arises when the slot ID in the BLPackage is generated automatically, either because it was left blank or set to -1. To ensure that rollback proceeds successfully, provide a valid slot ID.

Workaround: Manually delete the adapter.

USP job taking long time for enrolling the VM 

This issue occurs because the Application Server was attempting to resolve the newly provisioned VM name before the VM (which was on DHCP) had the chance to register it with the DNS Server. This behavior lead to a "caching of name lookup" failure in the server's DNS Client Cache (Negative Caching in DNS). As a result, all further lookups for the VM name failed until the cached failure was discarded on expiry of the TTL.

Workaround: On the Windows based App Server, set TTL for Negative Caching to 0 (zero) by adding or modifying the MaxNegativeCacheTtl field in the Windows registry as described here: http://support.microsoft.com/kb/318803. Further discussion about this problem can be found here: http://social.technet.microsoft.com/Forums/en-US/winserverNIS/thread/6020b2eb-bcff-470d-a9b3-fa611947cfc5.

Cannot package VMware Licensed Feature properties

The Licensed Feature properties VMotion, VMware DRS, and VMware HA cannot be packaged into a BLPackage from the vCenter server.

Assets are packageable only if they can be deployed or if there is a child asset that can be deployed. Neither of those conditions are met by the licensed features.

Agent stops when executing large number is snapshots at the same time

When executing 50 snapshot instances simultaneously for a vCenter server, the RSCD Agent may stop. This issue is associated with using the vmware-instances.nsh script to run Snapshot, Audit, and Compliance Jobs on Virtual Machine components that reside on a vCenter server or an ESX server.

To resolve this issue, when you create the Snapshot, Audit, and Compliance Jobs, set the Number of targets to process in parallel " to Limited, and set the number of targets to a number appropriate to the hardware that supports the Virtual Center server or ESX server. In general, a number smaller than five produces good results.

Error encountered after upgrading BSA app server.

After upgrading the BSA app server, ensure that you also upgrade the RSCD agent on the virtual server and the CO distributed. If this is not done you will encounter job errors.

Issue with configuration objects running on a Red Hat Linux

Certain virtualization configuration objects running on a Red Hat Linux computer might encounter the following error: any action that you try to perform through the TrueSight Server Automation Console for the management of the Citrix XenServer server assets results in an error message, "Unable to start CO Server."

This issue arises when resource usage by the Citrix XenServer configuration object exceeds the defined OS limitations for the allocation of shared memory segments or message queues. See Troubleshooting-virtualization-configuration-objects-issues for details.