Powering down the storage head/controller

Performing a graceful shutdown ensures that data is saved and not corrupted, and that resources are assigned to the other controller in the storage head cluster. This is the preferred method for powering down a controller for component replacement.

1. Ensure that Ethernet cables are connected from your network to the NET-0 port on the back of each server.

2. Direct your web browser to the server to be serviced by using either the IP address or host name assigned to the NET-0 port as follows: https://ipaddress:215.

3. Log in as root, using the system-wide Oracle Virtual Compute Appliance password.

4. Go to Maintenance > Hardware.

5. Click the Show Details link for the server.

6. Click the Power icon for the server and select Power off from the pull-down list.

1. Press and quickly release the Power button on the front panel.

   This action causes an orderly shutdown of the operating system, and the server enters the standby power mode.

2. If the server did not respond or you need a more immediate shutdown, press and hold the Power button for four seconds.

   This forces the main power off and enters the standby power mode immediately. When the main power is off, the Power/OK LED on the front panel begins flashing, indicating that the server is in standby power mode.

Powering down the disk shelf

Do not remove a component if you do not have an immediate replacement. The disk shelf must not be operated without all components in place. Powering down or removing all SAS chains from a disk shelf will cause the controllers to panic to prevent data loss. To avoid this, shut down the controllers before decommissioning the shelf.

1. Stop all input and output to and from the disk shelf.

2. Wait approximately two minutes until all disk activity indicators have stopped flashing.

3. Place the power supply on/off switches to the "O" off position.

4. Disconnect the power cords from the external power source.

Powering on the storage appliance

The disk shelf must not be operated without all components in place.

1. Reconnect the disk shelf power and data cables you removed to service a component.

2. Place the power supply on/off switches on the disk shelf to the "I" on position.

3. Wait several minutes until the boot process is complete, at which time the Power LED should be solid green.

4. Connect the storage head power and data cables you removed to service a component.

5. Power on the server by pressing the Power button on the front panel.

   If you are not physically located at the system, use either of these ILOM methods instead:

   • Log in to the Oracle ILOM web interface.

     In the left pane, click Host Management > Power Control, and in the Actions list click Power On.

   • Log in to the Oracle ILOM command-line interface (CLI).

     At the CLI prompt, type the following command: start /System.

6. Wait approximately two minutes until the power-on self-test (POST) code checkpoint tests have completed, and the Power/OK LED on the front panel lights and remains lit.

7. If you performed a graceful shutdown earlier, return resources to the server that was just serviced.

   1. Log into the web UI for the server that was not serviced.

   2. Go to Configuration > Cluster.

   3. Click Failback.

   Note

   For information about configuring the clustered servers and attached disk shelves, see the “Oracle ZFS Storage System Administration Guide” for the appropriate software release.