Configuring a storage cluster

Configure a storage cluster by using the storage nodes for which the setup of each storage node is complete. Use the BMC remote console to perform this procedure.

Notes on the procedure

Perform this procedure only for one storage node of your choice (referred to as representative storage node hereafter). Do not change the target storage node during the procedure.
Communication is performed from the representative storage node to other storage nodes to configure a storage cluster.
For details about how to operate BMC (such as how to display and operate the BMC web interface window and how to operate the BMC remote console), see the BMC User Guide provided by the vendor of the physical server.

About the console interface screen

For the display items and contents of the console interface screen, see the VSP One SDS Block Bare Metal Setup and Configuration.

CAUTION:

If an event that should be notified to the user occurs, a message and message ID are shown in the console interface.

For details about action to be taken when this message appears, see [Solution] of the corresponding message description in the VSP One SDS Block and SDS Cloud Message Reference.

Transfer the VSSB configuration file containing the storage cluster configuration information from the controller node to the representative storage node by using SFTP.

The following indicates the information required for SFTP transfer and an example of a transfer command.

Item	Description
Username	setup
Password	Password of the setup user that was set during setup of each storage node
Port number	10022
Destination IP address	IP address of the control port of the representative storage node (the item corresponding to "ControlNWIPv4" in the VSSB configuration file)
Transfer destination directory	/config

Note:

When the procedure described in Setting up each storage node is complete, control ports of the storage nodes or spare nodes are enabled, enabling communication confirmation by Ping from the controller node to control ports of the storage nodes or spare nodes.

Command example:

For Windows:

> sftp -P 10022 setup@<IP-address-of-control-port-of-representative-storage-node>
sftp> cd config
sftp> put SystemConfigurationFile.csv

For Red Hat Enterprise Linux or Debian:

$ sftp -P 10022 setup@<IP-address-of-control-port-of-representative-storage-node>
sftp> cd config
sftp> put SystemConfigurationFile.csv

Note:

When you connect to a storage node using SFTP, the node allows only minimum remote operations for directories and files from the SFTP client.

In interactive mode after the storage node is connected, only the following sftp commands can be used for remote directories and files:
- pwd
- ls
- cd
- get
- reget
- put
- reput
- rm

When the controller node connects to the representative storage node by SFTP connection, the following message might be output.

Output message*:

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
@    WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED!     @
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
IT IS POSSIBLE THAT SOMEONE IS DOING SOMETHING NASTY!
Someone could be eavesdropping on you right now (man-in-the-middle attack)!
It is also possible that a host key has just been changed.
The fingerprint for the ECDSA key sent by the remote host is
xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx.
Please contact your system administrator.
Add correct host key in /home/user/.ssh/known_hosts to get rid of this message.
Offending ECDSA key in /home/user/.ssh/known_hosts:xxxx
  remove with: ssh-keygen -f "/home/user/.ssh/known_hosts" -R xxx.xxx.xxx.xxx
ECDSA host key for xxx.xxx.xxx.xxx has changed and you have requested strict checking.
Host key verification failed.

* This is an example of the message that is output when the sftp command is run. The message that is output might differ depending on the SFTP client that you use.

This message notification occurs when the SFTP client detects that the controller node in use has previously connected to another physical server of the same IP address. Take action according to the message output by the SFTP client, and then retry SFTP connection.

Log in to the BMC web interface window of the representative storage node from the controller node, and then connect to the BMC remote console.

When the Keyboard layout configuration screen appears, select Login. (Use the up arrow (↑) and down arrow (↓) keys to select it, and then press the Enter key.)

The login screen for the setup user is displayed.
For Password, enter the password of the setup user set in the procedure in Setting up each storage node, and then press the Enter key.

If login is successful, the Top menu screen is displayed.
In the Top menu screen, select Execute setup.

The Execute setup (Caution) screen appears.
On the Execute setup (Caution) screen, read the displayed cautions and select OK if there are no problems.

The Execute setup (End User License Agreement) screen appears.

Note:
To return to the Top menu screen, select Back to top, and then select OK.
In the Execute setup (End User License Agreement) screen, select Submit.

The Execute setup (Progress and result) screen appears.

Note:
To return to the Top menu screen, select Back to top, and then select OK.
On the Execute setup (Progress and result) screen, verify the following:
1. Verify that the storage cluster configuration process has started.
  
  The progress of the storage cluster configuration process is then displayed.
2. When processing advances to Stage 5/8 in the Execute setup (Progress and result) screen, verify that all storage nodes restart once.
  
  After the restart, wait until the Keyboard layout configuration screen appears on the BMC remote console of the representative storage node.
  
  After the progress of Stage 5/8 is displayed on the Execute setup (Progress and result) screen, it takes approximately 15 minutes until the Keyboard layout configuration screen appears after restart.
  
  CAUTION:
  
  When setup of a storage cluster has started, do not use the console interface of any storage nodes other than the representative storage node until setup ends. If you do so, setup might be unsuccessful.
  
  If configuration of a storage cluster is not successful, take action according to the VSP One SDS Block and SDS Cloud Troubleshooting Reference.
When the Keyboard layout configuration screen appears, select Login.

The setup user login screen appears.
Enter the password of the setup user for Password, and then press the Enter key.

If login is successful, the Execute setup (Progress and result) screen is displayed.

Note:
In this procedure, the Top menu screen is not displayed when login is successful.

If the Top menu screen is displayed upon successful login, the storage node you are operating might not be the representative storage node. If the Top menu screen is displayed, select Logout to log out, and then log in to the correct representative storage node.
On the Execute setup (progress and result) screen, verify that the process of building the storage cluster is ongoing.

The progress of the storage cluster configuration process is then displayed.

After the processing proceeds to Stage 6/8 in the Execute setup (Progress and result) screen, it takes about 25 minutes for storage cluster setup processing to complete.

This processing also confirms that startup of each storage node is complete. This confirmation needs about a maximum of 60 minutes. If there are storage nodes for which startup of physical nodes takes time, the processing might need a longer time.
CAUTION:
During setup of a storage cluster, if the KARS23760-W or KARS23761-W message is output on the Execute setup (Progress and result) screen, unreachable paths might be included in the networks of the control ports or internode ports that have a teaming configuration.

Because of the teaming configuration, the system can operate as a storage cluster even if some paths cannot be connected. However, for the network paths mentioned in either message, check the points regarding network connections and settings described in First points to confirm in the VSP One SDS Block and SDS Cloud Troubleshooting Reference and take a necessary action if a problem is found. After you address the problem, if you want to check the connection of control ports or internode ports in the VSP One SDS Block environment, perform either of the following:

(Even though an event log indicating restoration of the link state, such as KARS08304-I, KARS08305-I, KARS08306-I, KARS08313-I, KARS08314-I, and KARS08315-I, is output after the KARS23760-W event log or the KARS23761-W event log is output, reachability is not guaranteed. When you have concern about reachability other than the link state, confirm connections by performing the following operation.)
- Recheck the connection (Reperform initial installation)
  
  To recheck the connection of individual storage clusters, reperform the procedure from installation of storage software.
  
  If configuration of the storage cluster ends successfully and both the KARS23760-W message and KARS23761-W message do not appear on the Execute setup (Progress and result) screen, there is no problem with the connection status of the control network and internode network.
- Recheck the connection (Replace storage nodes)
  
  To recheck the connection of individual storage nodes, replace, one by one, any storage node with a connection source port that failed during connection checking. Replacing storage nodes does not stop the storage cluster itself. If the storage node that needs replacing is not blocked, perform storage node maintenance blocking before replacing the node. For details about storage node replacement and storage node maintenance blocking, see the VSP One SDS Block and SDS Cloud System Administration.
  
  If replacement of the storage nodes ends successfully and both the KARS10914-W event log and KARS10915-W event log are not output, there is no problem with the connection status of the control network and internode network for the replaced storage nodes.
Note:
In addition to the time required for storage node startup, the following external factors might increase or decrease the processing time. If processing takes a long time, verify that there are no problems with the following items:
- Network communication status (communication status between the storage nodes)
- Server performance status (including performance degradation due to temperature rise of storage nodes)
- Server hardware configuration (such as storage node processor or memory capacity)
When the storage cluster configuration processing ends, "Cluster setup completed" appears. After confirming that "Cluster setup completed" appeared, select Logout.
The Keyboard layout configuration screen appears. This completes configuration of the storage cluster.

CAUTION:

If an error is displayed or if the "Cluster setup completed" message is not displayed within 25 minutes, reinstallation is not successfully completed. A possible cause of reinstallation failure is inappropriate hardware replacement. Verify the model name of the new hardware (according to the VSP One SDS Block Bare Metal Hardware Compatibility Reference), its installation location, and whether there is any failure. Then, if there is a problem with hardware replacement, replace the hardware again. After that, reperform the tasks from Installing the storage software (for Hitachi Advanced Server HA800 series, HA800 G2 series, and HA800 G3 series) or Installing the storage software (for physical servers other than Hitachi Advanced Server HA800 series, HA800 G2 series, and HA800 G3 series).
Note:
You can no longer perform the following operations when setup of a storage cluster is complete:
- Login to the console interface as the setup user
- Connection to storage nodes via SFTP
The menu that is displayed when you log in to the console interface also changes. For details, see Operating the console interface in the VSP One SDS Block and SDS Cloud System Administration.