How to replace windows 2003 Server in the event of failure

How to replace a Windows 2003 Server in the Event of a Failure

Purpose

This article provides steps to replace a server in the event of a failure.

Resolution

Note: Read the entire process prior to attempting this procedure.

Initial Assumptions
  • The Operating System is installed on the C drive.
  • vCenter Server Heartbeat is installed on the C drive.
  • vCenter Server Heartbeat is installed in a LAN environment.
  • The Secondary server will be replaced.
  • The replacement server meets the documented prerequisites for vCenter Server Heartbeat installation (for example, ACPI compliant with the remaining server). For more information, see Prerequisites in the vCenter Server Heartbeat Reference Guide.
Before You Begin
  • Determine the identity of the server that needs to be replaced. If the server's identity is Primary, refer to the Replacing Primary Server section of this article.
  • Open the vCenter Server Heartbeat Console and go to the Filters section. Note in a file or on a piece of paper all of the replication filters.
  • Read the entire contents of this article.
Procedure
On the Primary Server:
  1. Shutdown vCenter Server Heartbeat and stop all protected applications when asked.
  2. Set Neverfail Server R2 to Manual using the Windows Service Control Manager.
  3. Open vCenter Server Heartbeat Configure Server wizard.
    1. Go to the Machine tab.
    2. Change the Physical Hardware Identity to Secondary.
    3. Change the Current Role to Passive.
    4. Click Finish.

  4. Search the C:\ drive for *.dmp and delete all existing .dmp files. Ensure that they are emptied from the Recycle bin.
  5. Run NTBackup.
    1. Select the entire C:\ drive and any other directories on a different drive that include program files of the protected application.
    2. Deselect any directories that are already part of the replication set.
    3. Click System State.
    4. Select a local location for the backup file.
    5. When this has finished, re-run the vCenter Server Heartbeat Configure Server wizard.
    6. Go to the Machine tab.
    7. Change the Physical Hardware Identity to Primary.
    8. Change the Current Role to Active.
    9. Click Finish.

  6. In the task bar, right-click on the vCenter Server Heartbeat Server icon and choose Start vCenter Server Heartbeat.
  7. Log into the vCenter Server Heartbeat Console. Select stop replication but leave the Protected Application running.
  8. Set Neverfail Server R2 to Automatic using the Windows Service Control Manager.
Replacing the Secondary Server
On the New Secondary Server:
  1. Copy the backup file from the Primary server.
  2. Unplug the principal (public) and channel cables from the server.
  3. Run NTBackup.
    1. Go to Tools > Options.
    2. Click the Restore tab.
    3. Select Always replace files on my computer and choose Apply.
    4. Click the Restore and Manage Media tab.
    5. Right-click File and choose Catalog File.
    6. Browse to the location and select the backup file that was copied from the Primary server.
    7. Expand the tree under the catalogued file and select all the backed up data (for example, C drive, System state).
    8. Click Start Restore and Ok to the warning message that follows.
    9. Click Ok to start the restore.
    10. Reboot the server after the restore process is complete.

  4. Let Plug and Play continue and reboot the server if asked to reboot.
  5. Check that the network cards IP settings are correct and ensure that the vCenter Server Heartbeat packet filter is enabled for the principal (public) connection and disabled for all channel connections.

    Note: It may be necessary to re-authenticate Windows.
  6. Plug in the principal (public) and channel network cables.
  7. Start replication.
  8. Set Neverfail Server R2 to Automatic using the Windows Service Control Manager.
Replacing the Primary Server
On the Secondary server:
  1. Start the Configure Server wizard.
  2. Click the Machine tab and change the Physical Hardware Identity of the server from Secondary to Primary.

    Warning: Do not perform this change in any other situation.
  3. Follow the procedure in Replacing the Secondary server described above.

    Note: In this case you will have your previously Secondary server running as the Primary server in the new pair.
Labels: ,

How to Resolve two active servers

How to Resolve two active servers


Symptoms

The occurrence of two active servers is not by design and, when detected, should be resolved immediately. When two identical active servers are live on the same network, VMware refers it as Split-brain syndrome.

Split-brain syndrome symptoms:

  • Both servers are running and in an active state. This is visible on the Taskbar icon as P / A (Primary and active) and S / A (Secondary and active).
  • An IP address conflict might be detected on a server pair running vCenter Server Heartbeat on the Principal (Public) IP address.
  • A name conflict might be detected on a server pair running vCenter Server Heartbeat. Typically, in a WAN environment, the Primary and Secondary servers connect to the network using different IP addresses. As a result, no IP address conflict occurs. However, if the servers are running with the same name then a name conflict may result. This only happens if both servers are visible to each other across the WAN.
  • Clients cannot connect to the server running vCenter Server Heartbeat.

Purpose

This article provides information about the symptoms and resolutions for two active servers.

Resolution

The most common causes of two active servers (Split-brain syndrome):

  • Loss of the VMware Channel connection (most common in a WAN environment).
  • The active server is too busy to respond to heartbeats.
  • Misconfiguration of vCenter Server Heartbeat software.

It is important to identify the cause of Split-brain syndrome and resolve it to prevent this issue from reoccurring.

When Split-brain syndrome occurs, the server with the most up-to-date data must be identified.

Note: If the wrong server is identified, it can result in data loss. Care should be taken to reinstate the correct server.

The following can help identify the server with the most up-to-date data:

  1. Check the date and time of files on both servers. The most up-to-date server will be made the active server.
  2. From a client PC on a LAN, run 'nbtstat -A 192.168.1.1' where the IP address should be the principal (public) IP address of your server. This can help identify the MAC address of the server currently visible to client machines.

Note: If both active servers have been servicing clients, perhaps at different WAN locations, only one server can be made active. Both servers contain recent data, which cannot be merged using vCenter Server Heartbeat. One server must be made active and one server passive to restart replication. Once replication is restarted, all data on the passive server is overwritten by the data on the active server. It may be possible to extract the up-to-date data manually from the passive server prior to restarting replication. Consult the Microsoft Knowledge Base for information regarding tools that can be used for this purpose. For additional information, please contact VMware Support.

To resolve two active servers:

  1. Identify the server with the most up-to-date data or the server you want to make active.
  2. Shutdown vCenter Server Heartbeat on both servers (if it is running).
  3. On the server you want to make passive, right-click the Taskbar icon, and select Server Configuration wizard.
  4. Select the Machine tab and set the server role to passive. Do not change the Identity of the server.
  5. Click Finish and reboot this server.
  6. Start vCenter Server Heartbeat (if required) and verify that the Taskbar icon reflects P / - (Primary and Passive) or S / - (Secondary and Passive).
  7. On the active server, right-click the Taskbar icon and select Server Configuration wizard.
  8. Select the Machine tab and verify the server role is set to active. Do not change the Identity of the server.
  9. Click Finish and reboot this server. As the server restarts, it connects to the passive server and starts replication. Data on the passive server is overwritten by data on the active server.
  10. Start vCenter Server Heartbeat (if required) and verify that the Taskbar icon reflects P / A (Primary and active) or S / A (Secondary and active).
  11. Log in to the vCenter Server Heartbeat Console.
  12. Verify that the servers have connected and replication has started.
Labels: ,

How to replace Passive Servers

How to Resolve two passive servers


Symptoms

If you have two passive servers, users are unable to connect to protected applications. This issue must be resolved immediately.

If alerts are configured, you receive notification that replication is not functioning properly.

This is the result of:

  • A sudden failure of the active server (for example, unexpected termination of the Neverfail Server R2 service, a transient power failure, a server reset triggered from the 'Power' or 'Reset' button, or any other unclean shutdown). Following an unclean shutdown, an active server automatically assumes the passive role to isolate itself from the network until the failure can be investigated.
  • A failure of the active server before completion of the handshake that establishes the VMware Channel. The passive server cannot detect that the active server is not responding when the failure occurs - no channel connection was established. The active server may suffer a transient failure and the passive server cannot respond by failing over into the active role leaving both servers in the passive role.
  • Both Primary and Secondary server experience a power outage simultaneously (for example, because they are using the same power source and neither is attached to a UPS). A failover is not be possible. When the servers are rebooted, each displays the error message:

    Cannot start replication because previous run did not shutdown properly. Check configuration

If you attempt to start vCenter Server Heartbeat without reconfiguring one server in the pair as active, Heartbeat responds with the warning:

[U16] Serious configuration mismatch between the two servers. Please reconfigure so there is one and only one Primary, and one and only one Active.

Purpose

This article provides steps to resolve a situation where two passive servers exist.

Resolution

To resolve two passive servers:
  1. Determine which server should be active.
  2. If vCenter Server Heartbeat is running on either server, shut it down. Leave any protected applications running on the server you want to make active.
  3. On the server you want to make active, open the Configure Server wizard.
  4. Click the Machine tab, select the active role, and click Finished. Do NOT change the Identity.
  5. Save the changes and exit the wizard.
  6. On the server you want to make passive, open the Configure Server wizard, and confirm that the role is passive. Do NOT change the Identity.
  7. Exit the wizard.
  8. Reboot both servers to ensure that all protected application services are stopped on the passive server and started on the active server.
  9. Start vCenter Server Heartbeat on both servers.
Labels: ,
Subscribe to: Posts (Atom)