Managing Cisco UCS Servers

This chapter contains the following sections:

Server Management

You can manage and monitor all blade and rack-mount servers in a Cisco UCS domain through Cisco UCS Director.

For information about how to select servers for management, see Selective Server Management.

Powering On a Server

Procedure

Step 1

Choose Physical > Compute.

Step 2

On Compute page, choose the pod that includes the Cisco UCS Manager account.

Step 3

Click UCS Servers.

Step 4

Click the row for the server that you want to power on.
Step 5

From the More Actions drop-down menu, choose Power On.

Step 6

Click Submit.

Powering Off a Server

Procedure

Step 1

Choose Physical > Compute.

Step 2

On Compute page, choose the pod that includes the Cisco UCS Manager account.

Step 3

Click UCS Servers.

Step 4

Click the row for the server that you want to power on.
Step 5

From the More Actions drop-down menu, choosePower Off.

Step 6

Click Submit.

Launching the KVM Console for a Server

Procedure

Step 1

Choose Physical > Compute.

Step 2

On Compute page, choose the pod that includes the Cisco UCS Manager account.

Step 3

Click UCS Servers.

Step 4

Click the row for the server for which you want to start the KVM console.
Step 5

From the More Actions drop-down menu, choose Launch KVM Console.

Step 6

Click Submit.

Cisco UCS Director downloads the kvm.jnlp file.

Step 7

Double-click on the kvm.jnlp file in your downloads folder.

The KVM Console opens in a separate window.

If you do not have the required Java Runtime Environment (JRE) installed, click More Info in the dialog box and follow the instructions to download and install the JRE.

Accessing a Server Directly using the KVM Console

You can access a UCS server directly using the KVM console.

Procedure

Step 1

Choose Physical > Compute.

Step 2

On Compute page, choose the pod that includes the Cisco UCS Manager account.

Step 3

Click UCS Servers.

Step 4

Click the for the server that you want to access directly using the KVM console.

Step 5

From the More Actions drop-down menu, choose KVM Direct Access.

Step 6

Click Submit.

The KVM Console opens in a separate window.

Step 7

Enter the user name and password, and select a domain.

Step 8

Click Launch KVM.

Requesting Inventory Collection for a Server

Procedure

Step 1

On the menu bar, choose Physical > Compute.

Step 2

Expand the pod and click Cisco UCS Manager account

Step 3

Click UCS Servers.

Step 4

Click the row in the table for the server for which you want to request inventory collection.
Step 5

Click Request Inventory Collection.

Step 6

Click Submit.

Issuing a Diagnostic Interrupt for a Server

Procedure

Step 1

Choose Physical > Compute.

Step 2

On Compute page, choose the pod that includes the Cisco UCS Manager account.

Step 3

Click UCS Servers.

Step 4

Click the row or the server for which you want to issue a diagnostic interrupt.
Step 5

From the More Actions drop-down menu, choose Server Maintenance.

Step 6

On the Server Maintenance screen, do the following:

  1. From the Server Maintenance drop-down list, choose Diagnostic Interrupt.

  2. Click Yes.

A Non Makeable Interrupt (NMI) is issued to the BIOS or operating system from the Cisco Integrated Management Controller (CIMC). This action creates a core dump or stack trace, depending on the operating system installed on the server.

Resetting a Server

Procedure

Step 1

Choose Physical > Compute.

Step 2

On Compute page, choose the pod that includes the Cisco UCS Manager account.

Step 3

Click UCS Servers.

Step 4

Click the row for the server that you want to reset.
Step 5

From the More Actions drop-down menu, choose Reset.

Step 6

Click Submit.

Reacknowledging a Server

Perform the following procedure if you need to have Cisco UCS Manager rediscover the server and all endpoints in the server. For example, you can use this procedure if a server is stuck in an unexpected state, such as the discovery state.

Procedure

Step 1

Choose Physical > Compute.

Step 2

On Compute page, choose the pod that includes the Cisco UCS Manager account.

Step 3

Click UCS Servers.

Step 4

Click the row for the server that you want to re-acknowledge.
Step 5

From the More Actions drop-down menu, choose Server Maintenance.

Step 6

On the Server Maintenance screen, do the following:

  1. From the Server Maintenance drop-down list, choose Re-acknowledge.

  2. Click Yes.

Cisco UCS Director sends a request to have Cisco UCS Manager disconnect the server and then build the connections between the server and the fabric interconnect or fabric interconnects in the system. The acknowledgment may take several minutes to complete.

Decommissioning a Server

Procedure

Step 1

Choose Physical > Compute.

Step 2

On Compute page, choose the pod that includes the Cisco UCS Manager account.

Step 3

Click UCS Servers.

Step 4

Click the row for the server that you want to decommission.
Step 5

From the More Actions drop-down menu, choose Server Maintenance.

Step 6

On the Server Maintenance screen, do the following:

  1. From the Server Maintenance drop-down list, choose Decommission.

  2. In the Reason field, enter the reason for decommissioning the server.

  3. Click Yes.

Cisco UCS Director shuts down the server, removes it from the Cisco UCS configuration, and adds it to the Decommissioned Servers tab.