Launching soon: The world's first vendor agnostic bug scrubLearn more & join waitlist
OPERATIONAL DEFECT DATABASE
...
...
This article provides steps to recover VMware Horizon from backup.Beginning with Horizon 7, it is possible to recover the Horizon Global AD LDS database from the backup. This database stores configuration data for Cloud Pod Architecture (CPA) federations. Important: This is also applicable to Horizon 7 We recommend you migrate from Horizon 7 to Horizon 8 as soon as possible. Horizon 8 offers numerous improvements in performance, scale, and experience for both Administrators and end users. Please see Ensuring a successful migration from Horizon 7 to Horizon 8 (89840) for additional advice.
This article is divided into four procedures. Follow the procedure that best addresses your scenario. In all cases, refer to the Reconciliation*** section at the end. Restore Procedure for Site Recovery with Machine BackupsRestore Procedure for Site Recovery without Machine BackupsProcedure for Recovering from a Corrupted CPA Database SchemaProcedure for Replacing a Failed Connection Server Instance Procedure for Recovering from a Corrupted CPA Database Schema If the Global AD LDS database schema has become corrupted, global entitlements can be reinstated by importing the last backup file. Follow these steps: Dissolve the federation Nominate one pod as the “initial” pod. For all other pods, follow these steps: On one Connection Server in the pod: Open a command prompt and execute vdmUtil --authAs admin_user --authDomain admin_domain --authPassword admin_password –unjoin If this command fails, follow Recovering a Cloud Pod Architecture pod (2080253). On each Connection Server in the pod: Uninstall AD LDS instance ‘VMwareVDMDSG’ by following step 2 of Recovering a Cloud Pod Architecture pod (2080253). For the sole remaining pod, follow these steps: On one Connection Server in the pod: Open a command prompt and execute vdmUtil --authAs admin_user --authDomain admin_domain --authPassword admin_password --uninitialize If this fails because pods that are no longer accessible are still joined, they can be forcibly removed from the federation by executing vdmUtil --authAs admin_user --authDomain admin_domain --authPassword admin_password --ejectPod --pod pod_name If this also fails, or there are no other pods, follow Recovering a Cloud Pod Architecture pod (2080253). On each Connection Server in the pod: Uninstall AD LDS instance ‘VMwareVDMDSG’ by following step 2 of Recovering a Cloud Pod Architecture pod (2080253). Recreate the federation On one Connection Server in the “initial” pod, follow these steps: Log in to Horizon Console as a Global AdministratorSelect Horizon 7 Configuration > Cloud Pod Federation and Click the "Initialize" option to create a new CPA federationLogon to the Connection Server with an AD account with Horizon Administrator privilegesStop the Windows service ‘VMware Horizon Connection Server’Copy the Global AD LDS database backup file from backup to the local file systemIf the backup file is encrypted (this is normal), then open a command prompt and decrypt it using, vdmimport -d -p password -f VDMGlobalConfig.LDF > decrypted.LDF Note: If you do not remember the data recovery password, run this command without the -p option. The utility displays the password reminder and prompts you to enter the password. If the file cannot be decrypted, please note this password is not retrievable. Import the decrypted file to restore the Global configuration using vdmimport -g -f decrypted.LDF Start the Windows service ‘VMware Horizon Connection Server’Logon to Horizon Console and verify that the health dashboard is green Rejoin the federation For all other pods, follow these steps: Log in to Horizon Console as a Global AdministratorSelect Horizon 8 Configuration > Cloud Pod Federation and Click the "Join" option to rejoin the CPA federation Procedure for Replacing a Failed Connection Server Instance If a Connection Server fails, but other Connection Servers in the cluster or pod are operational, perform these steps: Remove the existing Connection Server entry. Open a command prompt on any other Connection Server and remove the database entry for the failed machine like this: vdmadmin -S -r -s server, where server is the name of the failed machine. Install the new Connection Server. Create a suitable VM or provision a physical machine as required.Install operating system and updates.Install Connection Server replica.If the file system of the failed machine is still accessible and there is a trust store or locked.properties file in place, copy it or them to the new machine. The location of these files is install_path\VMware\VMware View\Server\sslgateway\conf. Restart the Windows service ‘VMware Horizon Security Gateway Component’.If the installation is part of a Cloud Pod Architecture federation, then after this step, the Cloud Pod Architecture feature will be configured automatically on the new Connection Server instance. Remove the old Connection Server Shut down the failed instanceIf the machine is virtual, delete the VMIf the machine is physical and the hardware is fine, reformat its hard disk and re-use. *** Reconciliation After a restore of the Local AD LDS database, missing desktops may appear in Horizon Console if the following actions are taken after the backup but before a restore: An administrator deleted pools/desktops.A desktop pool was recomposed resulting in the removal of un-assigned desktops (spare virtual machines).Missing desktops/pools can be manually removed from the Horizon Console.Some automated desktops may become disassociated from their pools as a result of creating a pool between the time the backup took place and restore time. Horizon Administrators may be able to return them to use by cloning the linked clone desktop as a full clone desktop through vCenter Server, creating it as an Individual Desktop in Horizon Console and assigning such desktops to a specific user. After restoring vCenter Server, remove orphaned virtual machines pertaining to View Server Pools using the VMware Infrastructure Client. If orphaned virtual machines and templates appear as disconnected in VMware Infrastructure Client, restart the VirtualCenter Server service. Disconnected virtual machines may appear with the erroneous status of READY in Horizon Console.Contact VMware Support for instructions for removing orphaned sources and replicas under VMwareViewComposerReplicaFolder after a database restore. Some orphaned replicas and sources may be cleaned up automatically by Horizon Composer. Finding and removing unused replica virtual machines in the VMware Horizon View can assist with cleanup.Verify that the new environment is functional and review the steps of this procedure once more if you encounter problems. For further help, contact support by following this link: https://www.vmware.com/support/file-sr.html