Connect to Server operation

Use the Connect to Server operation to disconnect the IBM Storage Scale Client from the current IBM Storage Scale Server and connect to the server that is currently configured in the IBM® shared service for IBM Storage Scale.

Before you begin

If the IBM Storage Scale client is not authorized with the IBM Storage Scale server cluster's file system before this call, this operation only unmounts the file system. So, if the IBM Storage Scale client already mounted the file system that is managed by this server cluster, the file system is not unmounted.

If the client is not yet authorized with the shared file system by the cluster IBM Storage Scale server that the IBM shared service for IBM Storage Scale instance points to, this operation unmounts and then mounts the file system again. This case occurs when a new IBM Storage Scale Server instance is deployed and the IBM Storage Scale Shared Server instance is redeployed to point to that instance, or when a Passive takeover happened and the Passive cluster is now managing the file system. To ensure that the file system can be unmounted successfully, the following conditions must be met:

  • There should be no open files in the file system.
  • There should be no processes using a subdirectory in the file system as the current working directory.
  • The system must be quiesced while the operation is in progress.

Refer to your IBM Storage Scale product documentation for information on lsof and fuser system utilities, which might be helpful in identifying and stopping processes that reference the file system. Also refer to the mmumount command for additional information about unmounting the file system. It is your responsibility to ensure that the file system can be unmounted successfully.

Do not perform this operation on IBM Storage Scale Clients that are configured and deployed with the GPFS Client Install for off rack GPFS Server option.

This operation is available for IBM Storage Scale Clients that are configured to connect to a IBM Storage Scale Server by using the IBM Storage Scale Client policy. You can also perform this operation if you configured your IBM Storage Scale Client to connect to the IBM Storage Scale Server by using the IBM Storage Scale Client script packages, and if your instance is not a classic virtual system instance.

About this task

You should perform this operation on the IBM Storage Scale Client after any of the following configuration changes have been made to the IBM Storage Scale Server:
  • A new key is installed on the IBM Storage Scale Server.
  • The IBM Storage Scale key for the IBM Storage Scale Server is changed.
  • The IBM shared service for IBM Storage Scale was redeployed and points to a different IBM Storage Scale Server instance cluster. The client needs to authorize with the new server cluster.
The following high level scenario provides an example of when you would perform this operation:
  • You create and deploy a IBM Storage Scale Primary and IBM Storage Scale Mirror / Tiebreaker configuration.
  • You deploy an instance of the IBM shared service for IBM Storage Scale with information about the IBM Storage Scale Primary server.
  • You deploy a IBM Storage Scale Client which communicates through the shared service to the Primary server.
  • The Primary deployment later develops a problem and can no longer be used.
  • You run failover operations on the Mirror deployment, transforming it into a new Primary configuration.
  • You optionally create and deploy a new Mirror configuration and attach it to the new Primary configuration.
  • You delete the existing shared service instance.
  • You deploy a new instance of the IBM shared service for IBM Storage Scale with information about the new Primary server (for more information on this process see the Related link).
  • You run the Connect to GPFS Shared Service server or Connect to same GPFS server operation to establish communication between the IBM Storage Scale Client and the new IBM Storage Scale Primary server, through the IBM shared service for IBM Storage Scale or the IBM Storage Scale server that you specify by using the IBM Storage Scale Manager IP address and Client Key field.
Note: You might want to run this operation when a mirror or tiebreaker configuration is added or removed after the client is deployed. Otherwise, the Show Client Status operation will not display accurate information about the remote server cluster nodes.

Procedure

  1. Expand the Fundamentals section of the GPFSClient operations.
  2. Expand the Connect to Server node.
  3. Select an option from the Connect to list:
    Connect to GPFS Shared Service server
    Choose this option to reconnect to the IBM Storage Scale server that is configured in the IBM Storage Scale shared service.
    Same GPFS server
    Choose this option to connect to the same IBM Storage Scale server that you specified at deployment time.
    GPFS Server defined below
    Choose this option to configure a IBM Storage Scale server for the instance. Specify the IBM Storage Scale Manager IP address and Client Key for the IBM Storage Scale server that you want to use for the instance.
  4. Click Submit.
  5. Click Confirm when asked to validate your request.

Results

This operation performs the following tasks:
  1. Information about the current IBM Storage Scale Server connection is retrieved from the IBM shared service for IBM Storage Scale.
  2. IBM Storage Scale keys are exchanged between the Client IBM Storage Scale Cluster and the Server IBM Storage Scale Cluster.
  3. If the Client IBM Storage Scale Cluster is authorized to connect to the Server IBM Storage Scale Cluster, it is added to the list of clusters that can connect to the Server IBM Storage Scale Cluster.
  4. The Client IBM Storage Scale Cluster configuration is updated as needed with information about the Server IBM Storage Scale Cluster and its nodes.
  5. File systems on the Client IBM Storage Scale Cluster are unmounted and remounted back, only if the IBM Storage Scale client was not authorized with the IBM Storage Scale server cluster file system before this call.
A new entry is displayed under the Operation Execution Results list, with the name Connect to Server. When completed, the status of the operation changes from Active to Done, and the Return Value column displays a message that the connection to the server was successful, similar to the following example:
Web_Application-was.11402667490454.GPFSClient: Connect to Server: Success

You can view more details about the operation in theIBM Storage Scale Clientt trace log for the associated virtual machine. From the Log Viewer, expand the virtual machine name node, then the IWD Agent node, then the .../logs/<vm_name>.GPFSClient node, and open the trace.log file. Verify that there are no errors in this log.

If the Connect to Server operation fails, refer to the Related links for troubleshooting information regarding failing to update the remote file system information.