List of points to confirm (Bare metal)

Points to confirm	Description
Verify that the cables of the storage system and network peripheral devices are connected correctly.
	Simple errors such as disconnection or incorrect connection of network cables might often cause a failure. Verify that network cables are connected to correct network switches.
	When the network is redundant, verify that all the network cables are connected to the correct network switches. If a connection error persists after switching to the alternate path for failover, the network cable for the alternate path might not be connected to the correct network switch.
	When the VLAN function is used for a network switch, verify that the network cable is connected to the interface of the network switch for which VLAN is correctly set. For how to operate the network switch, see the hardware vendor documentation.
Verify that power is supplied to the storage system and network peripheral devices.
	If power to the storage system stops, you must restart the storage system after power supply is restored.
Verify that you performed the necessary tasks by using the correct procedure as described in the document.
	Verify the content of the notes and cautions described in the procedures again.
When you perform an operation from the REST API/CLI, verify that the access-destination address is correctly specified.
	Specify the correct IP address or host name for the storage cluster or individual storage node that you want to work with. For details on specifying the address in the CLI, see the VSP One SDS Block and SDS Cloud CLI Reference.
If you use an FQDN to specify the destination when you perform an operation from the REST API/CLI, verify that the FQDN can correctly resolve to an IP address.
	Verify that the DNS record is correctly set for the FQDN. When changing or setting configuration information, the new IP address of the control network must be set as a DNS record for the FQDN.
	Verify that the DNS server setting is completed.
	Verify the path to the DNS server for any failures.
	Verify that the DNS server is running.
When using REST API/CLI, verify that the parameters and other options are correctly specified.
	For more information, see the VSP One SDS Block and SDS Cloud REST API Reference or VSP One SDS Block and SDS Cloud CLI Reference.
When you perform an operation from the REST API/CLI, verify that the SSL/TLS communication settings are correctly specified.
	For more information, see the VSP One SDS Block System Administrator Guide.
Verify that the external servers accessed by the storage system are configured and operating normally.
	The storage system works in cooperation with multiple external servers. If the external servers are not set correctly, a failure might occur. Verify that the following external servers are configured and operating normally. External authentication (LDAP) server iSNS server SMTP server SNMP manager Syslog server DNS server NTP server For information about how to verify each external server setting, see the VSP One SDS Block System Administrator Guide.
	A user of the storage system might have to be registered with an external server, such as an LDAP server.
If KARS08202-E is output, verify that the NTP server is running.
	Verify that the settings of the NTP server to which the storage node is connected is correct. For details about the NTP server settings, see the VSP One SDS Block Bare Metal Setup and Configuration.
	When you specify the NTP server to which the storage node is connected with a host name (FQDN) and an event log is output, indicating an abnormal connection to the DNS server (KARS08200-E), see the event log, and then take action. Verify that the DNS record for resolution of the NTP server host name (FQDN) is correctly set. Also, verify that the NTP server is running normally and the network between the storage node and NTP server is normal. If any problems occur, resolve them.
If KARS08200-E is output, verify that the DNS server is running.
	Verify that the settings of the DNS server to which the storage node is connected is correct and the DNS server is running normally.
	Verify that the network between the storage node and DNS server is normal. If any problems occur, resolve them.
Verify that the control network, compute network, internode network, and BMC network that constitute VSP One SDS Block are all running normally.
	Use a network monitoring system or survey terminal that is independent of the VSP One SDS Block to check if there are any failures or performance degradation (network switch failure, link failure, increased delay, increased packet loss, connection at unexpected link speed and duplex operating values, and others) in the network. Continuous network degradation for five minutes or more affects operation of the storage system. I/O performance of the host might be degraded, or in the worst case, storage nodes might be blocked or other failures might occur. When temporarily connecting an investigation terminal to the internode network for fault isolation, be sure to disconnect after the fault isolation is completed, and verify that the network is closed with no connections other than the internode network port of the VSP One SDS Block storage node and the network monitoring system. For information on how to check the monitoring system and network switch of the network independent of the VSP One SDS Block, see the document of each vendor.
	When link-up state between the control port and internode port has continued for 30 seconds, the storage node judges that communication between them is possible. When the storage node judges that communication is possible, the following messages are output. KARS08304-I, KARS08305-I, KARS08306-I, KARS08313-I, KARS08314-I, KARS08315-I For detailed specifications, see Storage node requirements in the VSP One SDS Block Bare Metal Setup and Configuration. Note that if the storage node was blocked due to a link failure of internode network, the following messages indicating the link status recovery might not be output when you recover the link status between the time of failure occurrence and storage node recovery. KARS08313-I, KARS08314-I, KARS08315-I To verify that the internode network has recovered, see the link lamp on the server or the network switch to confirm the link status. Also, after performing a software update, if you see event logs output before the update, the port information in the messages of the following event logs indicating link state recovery might not be displayed correctly. KARS08305-I、KARS08306-I、KARS08314-I、KARS08315-I To obtain the latest port information, identify the storage node ID from the event log and follow the procedure as described in Obtaining Control Port Information Individually or Obtaining Internode Port Information Individually in the VSP One SDS Block System Administrator Guide.
	For details about how to view the BMC window, see the respective vendor's documentation. Also, the BMC window might not provide a link status. To see the status of the link, use the link lamp on the server or network switch.
	The link status of the control network, compute network, or internode network for a storage node that is either blocked or could not be blocked^* might not be correctly displayed by an operation performed from the REST API, CLI, or VSP One SDS Block Administrator. Verify the link status from, for example, the link lamp on the server or network switch. * If the storage node is either blocked or could not be blocked, the status will be one of the following: "TemporaryBlockage", "MaintenanceBlockage", "PersistentBlockage", "InstallationFailed", "TemporaryBlockageFailed", "MaintenanceBlockageFailed", "RemovalFailedAndTemporaryBlockage", "RemovalFailedAndMaintenanceBlockage", "RemovalFailedAndPersistentBlockage", "RemovalFailedAndMultipleFailures"
	If TCP retransmission increases during internode network, event log KARS08330-W is output. Possible causes of TCP retransmission are network failure, network performance degradation, or storage node (such as physical server or NIC) failure. Verify whether such phenomena occur. Particularly, verify that a network switch does not cause packet loss and connection is not at unexpected link speed or duplex.
	If you have installed a firewall on each network and are filtering using TCP / UDP port numbers, contact customer support and check if the TCP / UDP port number required for each communication is set to be allowed.
	If the link flap prevention feature of the network switch is enabled, verify that the network switch interface has not been shut down. If the network switch interface has been shut down by the link flap prevention feature, eliminate the cause of the link flapping, and then reactivate the network switch interface. For details about how to operate the network switch, see the hardware documentation.
	If the network is redundant and recovery processing such as switching to the redundant route is occurring due to a failure, check the log of the network switch, and verify that the time required for the recovery process meets the recovery time described in "Availability" of each network requirement in the VSP One SDS Block Bare Metal Setup and Configuration.
	In the following cases, there is a possibility that ARP packets used for connectivity verification cannot reach the path from the secondary port to the primary port (including inter-switch links if switches are configured redundantly). Verify whether there are any issues or failures along the path. When an event log indicating link state recovery is output, but no event log indicating failback is output even after a considerable amount of time has passed. When verifying the port information, both the primary and secondary ports are up, but the active port remains the secondary port.
Verify that the hardware of each storage node is operating normally.
	For details about how to verify that hardware is operating normally, see the hardware vendor documentation.
	Verify that power of each storage node is on. However, storage nodes that are no longer used (such as those that have been removed, replaced, have switched to spare nodes, or had their storage software uninstalled) do not need to be powered on.
	If the storage node must be powered down once to replace or reconfigure the hardware, the applicable storage node must be manually blocked for maintenance if the storage node is integrated into a storage cluster and has the Ready status. For details about the maintenance blockage of the storage node, see the VSP One SDS Block System Administrator Guide.
	For details about hardware maintenance including replacement of the server and its components, see the hardware vendor documentation or contact the customer support.
	If the response to the system drive access of a storage node is delayed, the message KARS08205-W is output. Verify whether there is any abnormality in the system drive, and take hardware measures if needed. In case of a temporary system drive access response delay, KARS08206-I is output 10 to 15 minutes after KARS08205-W is output, notifying that the system drive access response has become normal. KARS08206-I is not output if the system drive access response has become normal after the storage node is restarted or hardware is replaced. Therefore, determine that the system drive access response has become normal if KARS08205-W is not output even after 15 minutes of OS startup.
Verify that the hardware of each storage node is configured correctly.
	If the serial number and model name are not set, the storage node will fail to build. Check your hardware vendor's document for information on how to set up your hardware.
	If the storage node must be powered down once to perform reconfiguration or other changes to the hardware, the applicable storage node must be manually blocked for maintenance if the storage node is integrated into a storage cluster and has the Ready status. For details about the maintenance blockage of the storage node, see the VSP One SDS Block System Administrator Guide.
Verify that the hardware in use meets all the requirements described in the VSP One SDS Block Bare Metal Hardware Compatibility Reference.
	See the latest VSP One SDS Block Bare Metal Hardware Compatibility Reference. If the hardware in use does not meet the requirements, replace it with the hardware that meets the requirements.
	If the hardware in use no longer meets requirements because of software downgrade, collect the dump log files for all the storage nodes, and then contact customer support.
	If the storage node must be powered down once to replace the hardware, the applicable storage node must be manually blocked for maintenance if the storage node is integrated into a storage cluster and has the Ready status. For details about the maintenance blockage of the storage node, see the VSP One SDS Block System Administrator Guide.
Verify that the BIOS-related environment settings are correctly specified for each storage node.
	For details about how to specify the BIOS-related environment settings, see your server vendor documentation and Supported Server Models in the VSP One SDS Block Bare Metal Hardware Compatibility Reference.
	To reuse a storage node after its spare node has been switched, confirm that no one-time boot option is set. For details about the confirmation procedure, see Installing the storage software in the VSP One SDS Block Bare Metal Setup and Configuration.
	If the storage node must be powered down once to perform the BIOS-related environment settings, the applicable storage node must be manually blocked for maintenance if the storage node is integrated into a storage cluster and has the Ready status. For details about the maintenance blockage of the storage node, see the VSP One SDS Block System Administrator Guide.
Verify that the storage nodes that are no longer used have been disconnected from the control network, compute network, internode network, and BMC network that make up VSP One SDS Block.
	Removed storage nodes, replaced storage nodes, switched spare nodes, and storage nodes from which storage software has been uninstalled are no longer used. Verify that the physical servers of these storage nodes are disconnected from the networks. Also, verify that the OSs of these storage nodes are not started unexpectedly. For example, erase the data on the system disks or make the system disks not bootable. These storage nodes remaining connected with networks might cause problems. For example, the network settings remaining in the physical servers might cause IP address conflicts. In addition, event logs or audit logs related to storage nodes that are no longer in use might be output.
If you made port settings in the Network port configuration window according to Setting up each storage node in the VSP One SDS Block Bare Metal Setup and Configuration, verify that each port is configured correctly.
	In the following cases, verify that port settings (MAC addresses or WWNs) are as intended. If storage cluster configuration becomes unsuccessful during processing, compare A and C (shown later in this column) for all the storage nodes. If storage node addition becomes unsuccessful during processing, compare A and C (shown later in this column) for all the storage nodes to be added. If storage node replacement becomes unsuccessful during processing, compare A and C (shown later in this column) for all the storage nodes intended to be replaced. If an error related to a spare node network is output in the event log, compare A and C (shown later in this column) for the spare node for which the error is output. If a warning related to network redundancy is output in the event log, compare B and C (shown later in this column) for the storage node for which a warning is output. A. Setting contents of the port that is displayed by logging in to the BMC remote console and then selecting Show setting in the Top menu window. B. MAC address or WWN of the port of the storage node for which warning is output in VSP One SDS Block Administrator. C. Hitachi Advanced Server HA800 series, HA800 G2 series, HA800 G3 series, and HA800 G6 series: In iLO7, the MAC address under Ports that is among the network details displayed after selecting Host from the menu, navigate to Hardware > Network > select each network adapter. In iLO6 and earlier versions, MAC address or WWN of the port that is displayed by selecting System Information in the navigation tree of the iLO web interface window and then selecting Network tab. Other than Hitachi Advanced Server HA800 series, HA800 G2 series, HA800 G3 series, and HA800 G6 series: MAC address or WWN of the port displayed in the BMC web interface window.
Verify whether the VSP One SDS Block and compute node are correctly connected when KARS03228-I is output.
	If there is a blockage in a storage node within the storage cluster, follow the Cause and action taken in storage node maintenance (Bare metal) in the VSP One SDS Block System Administrator Guide to perform either the procedure Performing maintenance recovery for storage nodes (Bare metal) or Replacing storage nodes (Bare metal) to recover the applicable storage node.
	If maintenance operations are in progress, wait until they are completed.
	Perform the procedure as described in Obtaining a list of path information about compute nodes in the VSP One SDS Block System Administrator Guide to verify that paths are registered in a full mesh configuration for the compute nodes.
	If the compute port is iSCSI, see ALUA configuration guidelines in the VSP One SDS Block System Administrator Guide and configure ALUA on the compute nodes.
	If the compute port is NVMe/TCP, see ANA configuration guidelines in the VSP One SDS Block System Administrator Guide to configure ANA on the compute nodes.
	See the respective OS vendor's document and configure the system so that all paths from the compute nodes are recognized as online.
	If KARS03228-I continues to be output repeatedly after configuration, contact customer support.