Επεξεργασία

Κοινή χρήση μέσω

Recover a Configuration Manager site

  • Άρθρο

Applies to: Configuration Manager (current branch)

Run a Configuration Manager site recovery after a site fails or data loss occurs in the site database. Repairing and resynchronizing data are the core tasks of a site recovery and are required to prevent interruption of operations.

The sections in this article can help you recover a Configuration Manager site. To create a backup, see Backup for Configuration Manager.

Considerations before recovering a site

Important

This information applies only to site recovery scenarios. When you're upgrading your on-premises infrastructure and not actively recovering a failed site, review the information in the following articles:

Prepare the server hardware

Make sure existing configurations aren't present on the site server. Any previous configurations can cause conflicts during the site recovery process. Use one of the following options for the server hardware:

  • Use a new server, that meets the general and recovery requirements.

  • Format the disks, and reinstall the OS on the existing server. Make sure it meets the general and recovery requirements.

  • Reuse an existing server that you've cleaned

Use one of the following procedures to clean an existing server:

Clean an existing server for site server recovery only

  1. Delete SMS registry keys: HKLM\Software\Microsoft\SMS
  2. Delete any registry entries starting with SMS from HKLM\System\CurrentControlSet\Services. For example:
    • SMS_DISCOVERY_DATA_MANAGER
    • SMS_EXECUTIVE
    • SMS_INBOX_MONITOR
    • SMS_INVENTORY_DATA_LOADER
    • SMS_LAN_SENDER
    • SMS_MP_FILE_DISPATCH_MANAGER
    • SMS_SCHEDULER
    • SMS_SITE_BACKUP
    • SMS_SITE_COMPONENT_MANAGER
    • SMS_SITE_SQL_BACKUP
    • SMS_SITE_VSS_WRITER
    • SMS_SOFTWARE_METERING_PROCESSOR
    • SMS_STATE_SYSTEM
    • SMS_STATUS_MANAGER
    • SMS_WSUS_SYNC_MANAGER
    • SMSvcHost 3.0.0.0
    • SMSvcHost 4.0.0.0
  3. Uninstall the Configuration Manager console
  4. Restart the server
  5. Confirm that all of the above registry keys are deleted.

The server is now ready for the Configuration Manager restore procedure.

Clean an existing server for site database recovery only

  1. Back up the site database. Also back up any other supporting databases, like WSUS.
  2. Make sure to note the SQL Server name and instance name
  3. Manually delete the site database from the SQL Server
  4. Restart the SQL Server

The server is now ready for the Configuration Manager restore procedure.

Clean an existing server for full recovery

  1. Back up the site database. Also back up any other supporting databases, like WSUS.
  2. Make a copy of the content library

Warning

The following step - Uninstall the Configuration Manager site - should only be performed on a standalone Primary site, or a child Primary site that is unable to communicate over the network with the Central Administration Site (CAS). Uninstalling the site in a hierarchy results is the CAS losing the ability to communicate with that child primary and the restore process will fail. For child Primary sites, instead follow the Clean an existing server for site server recovery only steps above.

  1. Manually delete the site database from the SQL Server
  2. Uninstall the Configuration Manager site
  3. Manually delete the Configuration Manager installation folder, related registries and any other Configuration Manager folders
  4. Restart the server
  5. Restore the content library and other databases like WSUS.

The server is now ready for the Configuration Manager restore procedure.

Use a supported version and same edition of SQL Server

If possible, use the same version of SQL Server. However, it's supported to restore a database to a newer version.

Don't change the SQL Server edition. Restoring a site database from Standard edition to Enterprise edition isn't supported.

Other SQL Server configuration requirements:

  • SQL Server can't be set to single-user mode.
  • Make sure the MDF and LDF files are valid. When you recover a site, there's no check for the state of the files.

SQL Server Always On availability groups

If you use SQL Server Always On availability groups to host the site database, modify your recovery plans as described in Prepare to use SQL Server Always On.

Database replicas

After you restore a site database that you configured for database replicas, reconfigure each replica. Before you can use the database replicas, recreate both the publications and subscriptions.

Determine your recovery options

There are two main areas to consider for Configuration Manager primary site server and central administration site (CAS) recovery: the site server and the site database. The following sections can help you select the best options for your recovery scenario.

Note

When Configuration Manager setup detects an existing site on the server, you can start a site recovery, but the recovery options for the site server are limited. For example, if you run Setup on an existing site server, when you choose recovery, you can recover the site database server, but the option to recover the site server is disabled.

Site server recovery options

Start Configuration Manager setup from a copy of the CD.Latest folder that you created outside of the Configuration Manager installation folder.

  • If you run setup from the Start menu on the site server, the Recover a site option isn't available.

  • If you installed any updates from within the Configuration Manager console before you made your backup, you can't reinstall the site by using setup from the following locations:

    • Installation media
    • The Configuration Manager installation path

Then select the Recover a site option. You have the following recovery options for the failed site server:

Recover the site server using an existing backup

Use this option when you have a Configuration Manager backup of the site server from before the site failure. The site creates this backup as part of the Backup Site Server maintenance task. The site is reinstalled, and the site settings are configured based on the site that was backed up.

Reinstall the site server

Use this option when you don't have a backup of the site server. The site server is reinstalled, and you must specify the site settings as you would during an initial installation.

  • Use the same site code and site database name that you used when the failed site was first installed.

  • You can reinstall the site on a new computer that runs a new OS version.

  • The server must use the same hostname and fully qualified domain name (FQDN) of the original site server.

Site database recovery options

When you run Configuration Manager setup, you have the following recovery options for the site database:

Recover the site database using a backup set

Use this option when you have a Configuration Manager backup of the site database from before the database failure. The site creates this backup as part of the Backup Site Server maintenance task. In a hierarchy, when restoring a primary site, the recovery process retrieves from the CAS any changes made to the site database after the last backup. When restoring the CAS, the recovery process retrieves these changes from a reference primary site. When you recover the site database for a standalone primary site, you lose site changes after the last backup.

When you recover the site database for a site in a hierarchy, the recovery behavior is different for a CAS and primary site. The behavior is also different when the last backup is inside or outside of the SQL Server change tracking retention period. For more information, see the Site database recovery scenarios section in this article.

Note

If you select to restore the site database by using a backup set, but the site database already exists, the recovery fails.

Create a new database for this site

Use this option when you don't have a backup of the site database. In a hierarchy, the recovery process creates a new site database. When restoring a child primary site, it recovers the data by replicating from the CAS. When restoring the CAS, it replicates data from a reference primary site. This option isn't available when you're recovering a standalone primary site or a CAS that doesn't have primary sites.

Use a site database that has been manually recovered

Use this option when you've already recovered the Configuration Manager site database, but need to complete the recovery process.

  • Configuration Manager can recover the site database from any of the following processes:

    • The Configuration Manager backup maintenance task

    • A site database backup using Data Protection Manager (DPM)

    • Another backup process

      After you restore the site database by using a method outside Configuration Manager, run Setup, and select this option to complete the site database recovery.

      Note

      When you use DPM to back up your site database, use the DPM procedures to restore the site database to a specified location before you continue the restore process in Configuration Manager. For more information about DPM, see the Data Protection Manager documentation library.

  • In a hierarchy, when you recover a primary site database, the recovery process retrieves from the CAS any changes made to the site database after the last backup. When restoring the CAS, the recovery process retrieves these changes from a reference primary site. When you recover the site database for a standalone primary site, you lose site changes after the last backup.

Skip database recovery

Use this option when no data loss has occurred on the Configuration Manager site database server. This option is only valid when the site database is on a different computer than the site server that you're recovering.

SQL Server change tracking retention period

Configuration Manager enables change tracking for the site database in SQL Server. Change tracking lets Configuration Manager query for information about the changes made to database tables after a previous point in time. The retention period specifies how long change tracking information is kept. By default, the site database is configured to have a retention period of five days. When you recover a site database, the recovery process proceeds differently if your backup is inside or outside the retention period. For example, if your SQL Server fails, and your last backup is seven days old, it's outside the retention period.

For more information about SQL Server change tracking internals, see the following blog posts from the SQL Server team: Change Tracking Cleanup - part 1 and Change Tracking Cleanup - part 2.

Reinitialization of site or global data

The process to reinitialize site or global data replaces existing data in the site database with data from another site database. For example, when site ABC reinitializes data from site XYZ, the following steps occur:

  • The data is copied from site XYZ to site ABC.
  • The existing data for site XYZ is removed from the site database on site ABC.
  • The copied data from site XYZ is inserted into the site database for site ABC.

Example scenario 1: The primary site reinitializes the global data from the CAS

The recovery process removes the existing global data for the primary site in the primary site database and replaces the data with the global data copied from the CAS.

Example scenario 2: The CAS reinitializes the site data from a primary site

The recovery process removes the existing site data for that primary site in the CAS database. It replaces the data with the site data copied from the primary site. The site data for other primary sites isn't affected.

Site database recovery scenarios

After a site database is restored from a backup, Configuration Manager tries to restore the changes in site and global data after the last database backup. Configuration Manager starts the following actions after a site database is restored from backup:

Recovered site is a CAS

  • Database backup within change tracking retention period

    • Global data: The changes in global data after the backup are replicated from all primary sites.

    • Site data: The changes in site data after the backup are replicated from all primary sites.

  • Database backup older than change tracking retention period

    • Global data: The CAS reinitializes the global data from the reference primary site if you specify it. Then all other primary sites reinitialize the global data from the CAS. If you don't specify a reference site, all primary sites reinitialize the global data from the CAS. This data is what you restored from backup.

    • Site data: The CAS reinitializes the site data from each primary site.

Recovered site is a primary site

  • Database backup within change tracking retention period

    • Global data: The changes in global data after the backup are replicated from the CAS.

    • Site data: The CAS reinitializes the site data from the primary site. Changes after the backup are lost. Clients regenerate most data when they send information to the primary site.

  • Database backup older than change tracking retention period

    • Global data: The primary site reinitializes the global data from the CAS.

    • Site data: The CAS reinitializes the site data from the primary site. Changes after the backup are lost. Clients regenerate most data when they send information to the primary site.

Site recovery procedures

Use one of the following procedures to help you recover your site server and site database:

Start a site recovery in the setup wizard

  1. Copy the CD.Latest folder to a location outside the Configuration Manager installation folder. From the copy of the CD.Latest folder, run the Configuration Manager setup wizard.

  2. On the Getting Started page, select Recover a site, and then select Next.

  3. Complete the wizard by using the options that are appropriate for your site recovery.

    • During the recovery, setup identifies the SQL Server Service Broker (SSB) port used by the SQL Server. Don't change this port setting during recovery or data replication won't work properly after the recovery completes.

    • You can specify the original or a new path to use for the Configuration Manager installation in the setup wizard.

Start an unattended site recovery

  1. Prepare the unattended installation script for the options that you require for the site recovery. For more information, see Unattended site recovery.

  2. Run Configuration Manager setup by using the /script command-line option. For example, you create a setup initialization file ConfigMgrUnattend.ini. You save it in the C:\Temp directory of the computer on which you're running setup. Use the following command:

    setup.exe /script C:\temp\ConfigMgrUnattend.ini

Note

After you recover a CAS, replication of some site data from child sites can fail to be established. This data can include hardware inventory, software inventory, and status messages.

If this issue occurs, reinitialize the ConfigMgrDRSSiteQueue for database replication. Use SQL Server Manager to run the following query against the site database for the CAS:

IF EXISTS (SELECT * FROM sys.service_queues WHERE name = 'ConfigMgrDRSSiteQueue' AND is_receive_enabled = 0)  
 
ALTER QUEUE [dbo].[ConfigMgrDRSSiteQueue] WITH STATUS = ON

Post-recovery tasks

After you recover your site, there are several post-recovery tasks to consider before your site recovery is complete. Use the following sections to help you complete your site recovery process.

Reenter user account passwords

After a site server recovery, reenter the passwords for any user accounts in the site. These passwords are reset during the site recovery. The accounts are listed on the Finished page of the setup wizard after site recovery is completed. The list is also saved to C:\ConfigMgrPostRecoveryActions.html on the recovered site server.

Reenter user account passwords after site recovery

  1. Open the Configuration Manager console and connect to the recovered site.

  2. Go to the Administration workspace, expand Security, and then select Accounts.

  3. For each account, do the following steps to reenter the password:

    1. Select the account from the list identified after site recovery.

    2. Select Properties in the ribbon.

    3. On the General tab, select Set, and then reenter the password for the account.

    4. Select Verify, choose the appropriate data source for the selected user account, and then select Test connection. This step tests that the user account can connect to the data source, and verifies the credentials.

    5. Select OK to save the password changes, and then select OK to close the account properties page.

Reenter PXE passwords

  1. In the Configuration Manager console, go to the Administration workspace, and select the Distribution Points node. Any on-premises distribution point with Yes in the PXE column is enabled for PXE and may have a password to reenter.

  2. Select a PXE-enabled distribution point, and select Properties in the ribbon.

  3. Switch to the PXE tab.

  4. If the option to Require a password when computers use PXE is enabled, enter and confirm the password.

  5. Select OK to save and close the properties.

Repeat this process for any other PXE-enabled on-premises distribution point.

Reenter task sequence passwords

  1. In the Configuration Manager console, go to the Software Library workspace, expand Operating Systems, and select the Task Sequences node.

  2. Select a task sequence, and then in the ribbon, select Edit.

  3. Review the following steps for passwords to reenter:

    • Apply Windows Settings: If you enable and specify the local administrator password, reenter and confirm the password.

    • Apply Network Settings: For the account that has permission to join the domain, select Set. Enter and confirm the password, and then select Verify.

    • Capture Operating System Image: For the account used to access the destination, select Set. Enter and confirm the password, and then select Verify.

    • Connect to Network Folder: For the account used to connect a network folder, select Set. Enter and confirm the password, and then select Verify.

    • Enable BitLocker: If you use the key management option TPM and PIN, reenter the PIN.

    • Join Domain or Workgroup: For the account that has permission to join the domain, select Set. Enter and confirm the password, and then select Verify.

    • Run Command Line: If you use the option to Run this step as the following account, select Set. Enter and confirm the password, and then select Verify.

    • Run PowerShell Script: If you use the option to Run this step as the following account, select Set. Enter and confirm the password, and then select Verify.

Repeat this process for all task sequences.

Recreate bootable media and prestaged media in non-PKI environments

In non-PKI environments, self-signed certs in bootable media and prestaged media are based on the machine keys of the server where the media was created. For this reason, if the hardware changes or the OS is reinstalled as part of a recovery, any bootable media and prestaged media created on that server need to be recreated. For more information on how to create bootable media and prestaged media, see Create bootable media and Create prestaged media.

Reenter sideloading keys

After a site server recovery, reenter Windows sideloading keys specified for the site. These keys are reset during site recovery. After you reenter the sideloading keys, the site resets the count in the Activations used column for Windows sideloading keys.

For example, before the site failure the Total activations count shows as 100. The number of keys that devices have used, or Activations used, is 90. After the site recovery, the Total activations value still displays 100, but the Activations used column incorrectly displays 0. After 10 new devices use a sideloading key, there are no more sideloading keys, and the 11th device fails to apply a sideloading key.

Recreate Azure services

After site recovery, you may see the following error in the cloudmgr.log:

Index (zero-based) must be greater than or equal to zero

To resolve this issue, Renew the secret key for each Azure tenant connection.

Delete and recreate subscriptions for external notifications on the CAS

After you recover the CAS, you need to delete and recreate any subscriptions for external notifications. For more information, see External notifications.

Configure HTTPS for site system roles that use IIS

When you recover site systems that run IIS and you configured for HTTPS, reconfigure IIS to use the web server certificate.

Reinstall hotfixes

After a site recovery, you must reinstall any out-of-band hotfixes that were applied to the site server. After site recovery, view the list of the previously installed hotfixes on the Finished page of the setup wizard. This list is also saved to C:\ConfigMgrPostRecoveryActions.html on the recovered site server.

Recover custom reports

Some customers create custom reports in SQL Server Reporting Services. When this component fails, recover the reports from a backup of the report server. For more information about restoring your custom reports in Reporting Services, see Backup and Restore Operations for Reporting Services.

Recover content files

The site database tracks where the site server stores the content files. The content files themselves aren't backed up or restored as part of the backup and recovery process. To fully recover content files, restore the content library and package source files to the original location. There are several methods for recovering your content files. The easiest method is to restore the files from a file system backup of the site server.

If you don't have a file system backup for the package source files, manually copy or download them. This process is similar to when you originally created the package. Run the following query in SQL Server to find the package source location for all packages and applications: SELECT * FROM v_Package. Identify the package source site by looking at the first three characters of the package ID. For example, if the package ID is CEN00001, the site code for the source site is CEN. When you restore the package source files, they must be restored to the same location in which they were before the failure.

If you don't have a file system backup that includes the content library, you have the following restore options:

  • Import a prestaged content file: In a Configuration Manager hierarchy, you can create a prestaged content file with all packages and applications from another location. Then import the prestaged content file to recover the content library on the site server.

  • Update content: Configuration Manager copies the content from the package source to the content library. For this action to finish successfully, the package source files must be available in the original location. Do this action on each package and application.

Recover custom software updates

When you've included System Center Updates Publisher database files in your backup plan, you can recover the databases if the Updates Publisher computer fails. For more information about Updates Publisher, see System Center Updates Publisher.

Restore the Updates Publisher database

  1. Reinstall Updates Publisher on the recovered computer.

  2. Copy the database file Scupdb.sdf from your backup destination to %USERPROFILE%\AppData\Local\Microsoft\System Center Updates Publisher 2011\5.00.1727.0000\ on the computer that runs Updates Publisher.

  3. When more than one user runs Updates Publisher on the computer, copy each database file to the appropriate user profile location.

User State Migration data

As part of the state migration point properties, you specify the folders that store user state data. After you recover a state migration point, manually restore the user state data on the server. Restore it to the same folders that stored the data before the failure.

Regenerate the certificates for distribution points

After you restore a site, the distmgr.log might list the following entry for one or more distribution points: Failed to decrypt cert PFX data. This entry indicates that the distribution point certificate data can't be decrypted by the site. To resolve this issue, regenerate or reimport the certificate for affected distribution points. Use the Set-CMDistributionPoint PowerShell cmdlet.

Restore database encryption certificates

If you use SQL Server encryption for the entire database or for specific tables, you may need to restore the certificates after you restore the site database. For example, if you encrypt recovery data for BitLocker management. For more information, see Restore certificate for BitLocker management.

Recover a secondary site

Configuration Manager doesn't support the backup of the database at a secondary site, but does support recovery by reinstalling the secondary site. Secondary site recovery is required when a Configuration Manager secondary site fails.

Requirements

  • The server must meet all secondary site prerequisites and have appropriate security rights configured.

  • Use the same installation path that was used for the failed site.

  • Use a server with the same configuration as the failed server. This configuration includes its fully qualified domain name (FQDN).

  • The server must have the same SQL Server configuration as the failed site.

    • During a secondary site recovery, Configuration Manager doesn't install SQL Server Express if it's not already installed on the computer.

    • Use the same version of SQL Server and the same instance of SQL Server that you used for the secondary site database before the failure.

Procedure

Use the Recover Secondary Site action from the Sites node in the Configuration Manager console. Unlike with other types of sites, recovery for a secondary site doesn't use a backup file. This process reinstalls the secondary site files on the failed server. After the site reinstalls, the secondary site data is reinitialized from the parent primary site.

During the recovery process, Configuration Manager verifies if the content library exists on the secondary site server. It also checks that the appropriate content is available. The secondary site uses the existing content library, if it includes the appropriate content. Otherwise, to recover the content library of a secondary site, redistribute or prestage the content to the server.

When you have a distribution point that isn't on the secondary site server, you aren't required to reinstall the distribution point during a recovery of the secondary site. After the secondary site recovery, the site automatically synchronizes with the distribution point.

You can verify the status of the secondary site recovery by using the Show Install Status action from the Sites node in the Configuration Manager console.