How Do I Roll Back If Controller Expansion Is Paused Unexpectedly?

Prerequisites

• The operations in this section only apply to the original storage system. Before performing these operations, disconnect all cables from the new controller enclosure.
• If the 4 U storage system is expanded from 2 controllers to 4, 6, or 8 controllers, or from 6 controllers to 8 controllers, you must also remove the new controllers from the original controller enclosure.
• For details about how to restore controllers in a new 2 U controller enclosure, see Clearing the Controllers That Failed to Be Added.
• For details about how to restore controllers in a new 4 U controller enclosure, see Clearing the Controllers That Failed to Be Added.

Troubleshooting Flowchart

Procedure

1. Log in to any controller of the original storage system using the serial port or management network port and check whether the controller is in minisystem mode.
   • If yes, go to 2.
   • If no, run change user_mode current_mode user_mode=developer to enter the developer mode and run minisystem to switch to the minisystem mode. Then go to 2.

2. Run showbayconfig and check whether the value of CtrlTotal in the bay_confg.ini file of the controller has been changed to the total number of controllers after expansion.
   • If the bay_confg.ini file has been modified and you can access the CLI, go to 5.
   • If the bay_confg.ini file has been modified and you can only log in to the minisystem, go to 3.

     In this scenario, you must restart the controllers for the rollback to take effect, which will interrupt services. If you cannot tolerate the service interruption, go to 6.

   • If the bay_confg.ini file is not modified and you can access the CLI, you do not need to restore the original storage system. In this case, perform the controller expansion process again.
   • If the bay_confg.ini file is not modified and you can only log in to the minisystem, go to 6.

3. Log in to each controller of the original storage system and modify the bay_confg.ini file.
   1. Log in to the minisystem of each controller on the original storage system.
      • You must perform 3.b on each controller of the original storage system.
      • If you fail to log in to a controller (for example, controller B) using the serial port or management network port, log in to another controller in the controller enclosure and run the sshtoremoteExt SlotId command in minisystem mode (SlotId is the slot ID of controller B, that is, run the sshtoremoteExt 1 command) to the access controller B. SlotId is the controller's slot ID and its value ranges from 0 to N–1, where N indicates the maximum number of controllers that can be installed in the current controller enclosure.

        Storage: minisystem> sshtoremoteExt 1 
        Ssh to controller(SlotId=1) with admin...... 
        Warning: Permanently added '127.127.127.11' (ECDSA) to the list of known hosts.   
        
        Authorized users only. All activities may be monitored and reported. 
        password:

   2. Run the following commands in sequence to modify the bay_confg.ini file:
      1. Run os_write_bay_cfg.sh -c X.

         X indicates the number of controllers in the storage system before expansion.

      2. If the storage system was using the scale-out deployment before expansion, skip this step. Otherwise, run os_write_bay_cfg.sh -d.
      3. Run os_change_bayconfig.sh change.
      4. Run os_change_bayconfig.sh sync.

   3. Verify that you have modified the bay_confg.ini file for each controller of the original storage system. Then, go to 4.

4. Restart all controllers of the original storage system.

   You must run the restart command on each controller enclosure of the original storage system.

   1. Log in to any controller in each controller enclosure of the original storage system.
   2. Run the command for restarting the controller in minisystem mode.
      • The command will restart all controllers in the current controller enclosure.
      • For the 2 U storage systems, run the sys.sh resetengine -n command.
      • For the 4 U storage system:
        • If the controller enclosure has four controllers before expansion, run the sys.sh resetengine -n command.
        • If the controller enclosure has two controllers before expansion, run the sys.sh resetengine -p command.

      If a failure message is returned after the command is executed, log in to each controller of the original storage system, enter the minisystem mode, and run the rebootsys command to restart the controller.

   3. After the controller starts up, verify that it has been restored. If it is not restored, go to 6.

5. Run exit to exit the minisystem mode and run change controllers_expansion cancel controller_total=m in developer mode to roll back the expansion. m indicates the total number of controllers in the original storage system.

   After the command is executed successfully, the original storage system is restored and you can perform the controller expansion process again. If the storage system is not restored, go to 6.

6. Collect system logs and contact technical support for assistance.

   To collect system logs:

   • Log in to DeviceManager. If you can access the console, choose Settings > Export Data. Select All logs under System Log and click Log List. The system starts collecting logs.
   • If Failed to start the system is displayed when you log in to DeviceManager, click Download Log to download system logs.
   • If a controller fails to be powered on, log in to the controller and enter the minisystem mode. Then run os_backup_info.sh to collect logs and access the log path to obtain the logs.