Farm Wiring Issues and Solutions

During farm activation, various checks are made. When you run the /opt/terraspring/sbin/farm -l command, the farm state might display as NEW/ALLOCATED/30. This value indicates that the farm failed during the wiring phase. This section describes the possible causes and solutions to a farm wiring failure.

Moving a device into a VLAN fails. Moving a device into the image VLAN or farm VLAN during the wiring phase could fail in several ways.

Each failure has a unique solution:

• If the SNMP community string on the switch is not set correctly, moving a farm device into a VLAN might fail. To resolve this issue, change the public SNMP community string on the switch to match the admin password.

• The VLAN into which the device is being moved is not defined in the VLAN database of the chassis switch. To resolve this issue, update the VLAN database with the VLAN to which the farm device is being moved.

• The switch is not IP accessible from the Provisioning Server. To resolve this issue, resolve any networking issues that could cause IP connectivity to fail. Once the IP connectivity problem has been resolved, reissue the activation request using the farm command: /opt/terraspring/sbin/farm -af farm-id.

Unable to change the power state of a blade.

This problem has several possible causes and solutions:

• If a blade is marked faulty in the showplatform -v output on the chassis system controller (SC), the N1 Provisioning Server software will refuse to change the power state of the blade. A blade that has been marked faulty should be replaced.

• If the chassis SC login does not match the login and password known to the N1 Provisioning Server software, you will be unable to change the power state of the blade. In this case, you should update the chassis SC login to match the user name and password known to the software.

• During a power state change of a blade, the N1 Provisioning Server system needs to have IP connectivity to the chassis SC. If this connectivity is not granted, the power state change will fail.

• The chassis SC occasionally hangs. It does not respond or responds slowly to showplatform -v commands. In this case, you should turn off the power to the chassis and turn the power back on. Then, reactivate the farm using the following command: /opt/terraspring/sbin/farm -af farm-id.

• Finally, if you are trying to commit a power change to a device that is defined in the N1 Provisioning Server database but that has been physically removed from the chassis, the power command will fail. Issue a replace failed device request for the removed device in your farm and reactivate your farm.

Disk copy failed. If a disk copy fails, an error message appears in the Control Center window. Details about the error are in the debug log file /var/adm/tspr.debug.

The process of preparing a resource pool server for image copy is similar to the process executed during the final lab validation test during the installation. Refer to Installation Validation Issues and Solutions to troubleshoot a device that did not come alive within 300 seconds.

Image copies could also fail if the images requested to be copied to a resource pool server do not exist or are not in the READY state. To verify that the image you are trying to copy to a resource pool server is in the READY state, use the following command: /opt/terraspring/sbin/image -l image-id. To mark an image as ready to execute, use the following command: /opt/terraspring/sbin/imagesync --nosync image-id.

If the disk copy process starts successfully and later fails due to an interrupt, you might need to resolve networking problems within your environment.