Skip to main content
Outside service Partner
Hitachi Vantara Knowledge

Upgrading UCP Advisor

This document is for upgrading the UCP Advisor software on the management node instance from version 2.8.0 to version 2.9.0. Upgrading from versions prior to 2.8.0 is not supported. For information on upgrading versions prior to 2.8.0, see the relevant upgrade guide.

Preparing for the upgrade

Preparing for the upgrade will help ensure that the upgrade process goes smoothly and efficiently.

Before you go to the customer site:

  • Prepare a list of credentials for the system components you are upgrading.
  • Download the upgrade installation media and related documentation in case you do not have web access at the customer site.

At the customer site:

  • Ensure that the system you are upgrading is up and running with no active alerts, alarms, or unresolved issues.
  • Back up the UCP Advisor software configuration data.
  • Take snapshots of all of the management VMs for recovery.

Preparing a credentials list

Prepare a list of user credentials for each system component.

Work with the customer to provide a list of user credentials, either beforehand or available when you arrive on site. This list should include all user names and passwords for the current system components listed below, and will help save time when at the customer site.

  • vSphere (both vCenter root and vCenter SSO)
  • vCenter management port (root password for https://<vsphere_address>:5480)
  • Management hosts (root password for ESXi hosts hosting management VM URL https://<ManagementHost_address>/ui)
  • UCP Advisor controller VM
  • UCP Advisor Gateway VM
  • Hitachi Automation Director (if installed)
  • vRealize Orchestrator (if installed)
  • vRealize Log Insight (if installed)
  • Hi-Track Monitor
  • BMC IP and credential for compute servers

Verifying the UCP Advisor system

Prior to upgrading, verify that the UCP Advisor system is up and running with no active alerts or alarms, including its network and switches. All issues must be resolved before starting the upgrade. If you find any issues, discuss with the customer about how to resolve them before continuing with the upgrade.

  1. Verify compliance of the ESXi host profile. For details, see the VMware document, Checking compliance of a host profile.

  2. Verify that there are no active alarms in the vSphere Web Client. For details, see the VMware document, View and Edit Alarms.

  3. Create a list of all devices (network, Fibre Channel, compute, and storage) that are onboarded, online, and visible in UCP Advisor. Verify that there are no issues with any of the devices.

  4. Make sure the firmware versions of each device (switches, compute servers, storage) match the firmware version given in the release notes for that version.

  5. Fix any issues prior to the upgrade.

Downloading the UCP Advisor installation media ISO package

Download the media installation ISO package before going to the customer site because you might not have web access at the site where the upgrade is to be performed.

Procedure

  1. Log on to the Support portal at: https://support.hitachivantara.com.

  2. Click Product Downloads.

  3. In Find a Download, enter UCP Advisor.

  4. Under Available Software, click UCP Advisor.

  5. Download the 2.9.0 version of the UCP Advisor installation media.

  6. Verify that the media downloaded properly.

Backing up the UCP Advisor software configuration data

Back up the XML files used to configure the UCP Advisor software so that you can use them to restore the pre-upgrade configuration, if necessary. Verify with the customer if the XML files were modified from the standard installation version. If they were, you can use these backup files to reconfigure the software after the upgrade.

Procedure

  1. Log on to the UCP Advisor controller VM.

  2. Open an administrator CMD window.

  3. Run the following commands:

    md c:\temp\AdvisorConfigBackupFor2_8
    cd "C:\Program Files\Hitachi\UCP Advisor"
    xcopy /fcedy *.config c:\temp\AdvisorConfigBackupFor2_8
    For reference, an example of the XML configuration files are shown below.

Taking VM snapshots

Take snapshots of all management node VMs to enable recovery if needed.

  1. Log on to all management hosts, as the VMs may be spread across multiple management hosts.

  2. Shut down each VM in the following sequence. Allow each VM to shut down completely before shutting down the next one.

    1. UCP Advisor controller VM

    2. UCP Advisor Gateway VM

    3. Hitachi Automation Director, if present

    4. vRealize Orchestrator, if present

    5. vRealize Log Insight, if present

    6. vSphere VM

  3. In the Navigator pane, select Virtual Machines.

  4. Right-click the UCP Advisor controller VM, then select Snapshot > Manage Snapshots.

  5. Enter a name for the snapshot; for example, BeforeAdvisorUpgrade2_9.

  6. Click Take Snapshot.

  7. Repeat steps 4 through 6 to take a snapshot of each VM.

Upgrading the UCP ecosystem

Upgrade the software and VM components in the UCP ecosystem:

  • VMware vSphere
  • Hitachi Automation Director, if present
  • vRealize Orchestrator, if present
  • vRealize Log Insight, if present

Upgrading vSphere

Use the vCenter Server Appliance (VCSA) to upgrade the vSphere components to the level supported by the UCP Advisor upgrade version, as specified in the Release Notes.

Before you begin

Access the Release Notes for the upgrade version: https://knowledge.hds.com/Documents/Converged/UCP_Advisor.

For details on using this appliance, see Patching the vCenter Server Appliance and Platform Services Controller Appliance, available from VMware Docs at https://docs.vmware.com/en/VMware-vSphere/6.7/com.vmware.vcenter.upgrade.doc/GUID-043EF6BD-78F7-412F-837F-CBDF844F850C.html.

Procedure

  1. Bring the UCP Advisor controller VM online.

  2. Mount the VCSA Appliance Installer ISO package.

  3. Navigate to I:\vcsa-ui-installer\win32 and then run the installer.

  4. Select Upgrade and then follow the prompts to upgrade the VCSA appliance from version 6.5U2 to 6.7U1.

Next steps

Verify that vSphere updated correctly.

  1. On the UCP Advisor controller VM, log on to VMware Appliance Management at: https://<VCSA_address>:5480.
  2. Navigate to Access and then enable SSH login.
  3. Log on to the vSphere Web Client: https://<vSphereVM_address>/vsphere-client/?csp.
  4. Verify the updated version under Help > About VMware vSphere.

Upgrading optional software components

The following optional software components do not require upgrading for the current release of the UCP Advisor software:

  • vRealize Orchestrator, if present.
  • Hitachi Automation Director, if present.
  • vRealize Log Insight, if present.

Upgrading the UCP Advisor Gateway VM

Upgrade the software on the UCP Advisor Gateway VM, and then configure it.

Before you begin

Procedure

  1. Bring the UCP Advisor Gateway VM online.

  2. From a vCenter host, log on to the UCP Advisor Gateway VM console.

  3. Verify the currently installed version of the REST server:

    1. On the command line, enter puma_adm -pc to display the configuration information.

    2. Verify that the REST Server version is 7.0.0.3552.

  4. Verify the health of the Gateway VM: on the command line, enter puma_adm -ck | more.

  5. Start a Remote Desktop session, then log on to the UCP Advisor controller VM.

  6. Copy the UCP Advisor installation media ISO package you previously downloaded to C:\temp. If this directory does not exist, create it.

  7. Mount the ISO package listed above as F: drive.

  8. From an administrator command window, use the PSCP tool to copy the RPM file to the UCP Advisor Gateway VM. If you are prompted for an RSA key, enter y.

    "c:\Program Files\Hitachi\UCP Advisor\HivSphereConverged\InstallerUtil\putty\PSCP.EXE" -pw <UCP Advisor Gateway VM Password> F:\Software\AdvisorGatewayOVA\puma-7.1.0-1.el7.x86_64.rpm root@<UCP Advisor Gateway VM IP>:/root/

    In some cases, you may be prompted for the RSA keys.

    The server's host key is not cached in the registry. You have no guarantee that the server is the computer you think it is. The server's rsa2 key fingerprint is: ssh-rsa 2048 34:58:4b:9b:7c:cf:88:27:ae:52:93:1d:3a:03:fc:83 If you trust this host, enter "y" to add the key to PuTTY's cache and carry on connecting. If you want to carry on connecting just once, without adding the key to the cache, enter "n". If you do not trust this host, press Return to abandon the connection. Store key in cache? (y/n) y

    When copying is completed, the following line is displayed:

    puma-7.1.0-1.el7.x86_64.r | 465608 kB | 22186.1 kB/s | ETA: 00:00:00 | 100%
  9. From the UCP Advisor Gateway VM console, run the following command to update the software:

    puma_adm -r <rpm_file>

Next steps

Verify that the UCP Advisor Gateway VM upgraded correctly.

  1. Enter puma_adm -ck | more, then verify that the command device is attached. If any of the services are listed as FAIL, restart the puma service using the puma_adm -re command.
  2. Enter puma_adm -pc to display the configuration information, and then verify that the REST Server version matches the upgrade version as listed in the Release Notes.

Upgrading the UCP Advisor controller VM

Upgrade the UCP Advisor controller VM in multiple steps: verify that the UCP Advisor services are running, upgrade the third-party software, and then upgrade the UCP Advisor software.

Verifying UCP Advisor services

Verify that the UCP Advisor services are running on the UCP Advisor controller VM before upgrading the software.

Before you begin

Verify that the UCP Advisor controller VM is online.

Procedure

  1. Start a Remote Desktop session, and then log on to the UCP Advisor controller VM.

  2. Open the Services panel, and then verify that the following services are running:

    • Elasticsearch 5.5.0 (elasticsearch-service-x64)
    • Hitachi Syslog Agent for vRLI
    • Hitachi Baremetal
    • Hitachi UCP Advisor Ghost
    • Hitachi Hal
    • Hitachi UCP Advisor Hi-Track Web Service
    • Hitachi UCP Advisor Service Registry
    • Hitachi UCP Advisor Services
    • Hitachi UCP Advisor Sidecar
    • Hitachi UCP VI Service
    • Hitachi vRO Converge Web Service
    • Hitachi Watchdog Service

Uninstalling the UCP Advisor software

Uninstall the UCP Advisor software.

  1. Start a Remote Desktop session, and then log on to the UCP Advisor controller VM.

  2. From Control Panel > Programs and Features, uninstall the 2.8.0.x version of Hitachi Compute Advisor API.

  3. From Control Panel > Programs and Features, uninstall the 2.8.0.x version of Hitachi UCP Advisor.

    1. At the following prompt, click Yes:

      Do you want to completely remove the selected application and all of its features?
    2. In the InstallShield Wizard:

      • If you are upgrading the UCP Advisor software and you want to retain your existing UCP system information, click Skip.
      • If you want to completely uninstall the UCP Advisor software, including all UCP system information, enter the vCenter address and your vCenter SSO credentials, and then click Submit.
    3. If prompted, select Automatically close and attempt to restart applications, and then click OK.

    4. Click Finish.

  4. Manually restart the UCP Advisor controller VM.

  5. Open the Windows Services console and verify that all UCP Advisor services are removed. If any UCP Advisor services are still present, stop them and then use an administrator command prompt to remove them. For example:

    sc delete HiUcpServices
    sc delete HivROConvergeWebServiceHost
    sc delete UCPAdvisorHiTrack
  6. Delete the previous UCP Advisor installation folder. The default path is: C:\Program Files\Hitachi

Upgrading third-party software

UCP Advisor incorporates Elasticsearch and OpenJDK as third-party software components. This software must be upgraded before upgrading the UCP Advisor software.

The Elasticsearch database must be preserved before upgrading Java 8 to OpenJDK 11, and then the database must be restored after OpenJDK is running.

The third-party software is located in the UCP Advisor installation media ISO package at C:\temp. Mount this as drive F.

Preserving the Elasticsearch database

Preserve the Elasticsearch database and then uninstall Elasticsearch 5.5.0 before upgrading the Java software to OpenJDK.

Before you begin

Mount the UCP Advisor installation media ISO package in C:\temp as drive F.

Procedure

  1. Open a new administrator command window and then run the following script to preserve the Elasticsearch database:

    F:
    cd F:\Tools\ESBackup
    backup-es.bat
    NoteRun this script one time only. The script makes changes in the Elasticsearch service and running it multiple times will cause the service to fail.
  2. Navigate to the following folder and verify that it contains the indices folder.

    C:\Users\Administrator\Documents\esbackup

  3. Uninstall Elasticsearch 5.5.0.

    "C:\Program Files\elasticsearch-5.5.0\bin\elasticsearch-service.bat" remove
    The service 'elasticsearch-service-x64' has been removed.

Upgrading Java

Upgrade the Java software from Java 8 to OpenJDK 11.0.2.

Before you begin

  1. Preserve the Elasticsearch database.
  2. Uninstall Elasticsearch 5.5.0.

Procedure

  1. On the UCP Advisor controller VM, open Control Panel > Programs and Features, and then uninstall Java 8 Update 191.

  2. Browse to the following directory: F:\Software\OpenJDK11

  3. Copy folder jdk-11.0.2 to C:\Program Files\Java\

    Create a new Java folder if it is not already present.
  4. Update the path for system variable JAVA_HOME.

    1. Navigate to Control Panel > System and Security > System, and then click Advanced system settings.

    2. In the System properties window, click the Advanced tab, and then click Environment Variables.

    3. In the System variables pane, select JAVA_HOME, and then click Edit.

    4. Update the path to C:\Program Files\Java\jdk-11.0.2

    5. Click OK.

  5. Update the path for system variable Path.

    1. In the System variables pane, select Path, and then click Edit.

    2. In the Edit environment variable window, select C:\Program Files (x86)\Common Files\Oracle\Java\javapath, and then click Delete.

    3. Locate variable C:\Program Files\Java\jdk1.8.0_191\bin

      • If the variable is present, select it, and then update it to:

        C:\Program Files\Java\jdk-11.0.2\bin

      • If the variable is not present, click New, and then enter:

        C:\Program Files\Java\jdk-11.0.2\bin

    4. Click OK.

  6. Click OK.

  7. Restart the UCP Advisor controller VM.

  8. In an administrator command window, verify that OpenJDK is accessible.

Upgrading Elasticsearch

Upgrade and configure the Elasticsearch software.

Before you begin

Upgrade Java 8 to OpenJDK 11.0.2.

Procedure

  1. Copy the Elasticsearch 6.2.2 software to the following folder: C:\Program Files\ElasticSearch-6.2.2

    md "C:\Program Files\ElasticSearch-6.2.2"
    cd "C:\Program Files\ElasticSearch-6.2.2"
    xcopy /cedy F:\Software\elasticsearch-6.2.2
    
  2. Run the following command to install Elasticsearch 6.2.2.

    "C:\Program Files\ElasticSearch-6.2.2\bin\elasticsearch-service.bat" install
    Installing service	:"elasticsearch-service-x64"
    Using JAVA_HOME (64-bit): "C:\Program Files\Java\jdk-11.0.2"
    The service 'elasticsearch-service-x64' has been installed.
  3. Open Administrative Tools > Services, and then double-click the Elasticsearch service to open its properties panel.

  4. On the General tab, change the startup type to Automatic, and then click Start to start the service.

  5. Verify that the service is started.

  6. Click OK, and then close the Properties panel and Services window.

Restoring the Elasticsearch database

Restore the Elasticsearch database.

Before you begin

Upgrade the Elasticsearch software to version 6.6.2.

Procedure

  1. Open a new administrator command window, and then run the following script to restore the Elasticsearch database:

    F:
    cd F:\Tools\ESBackup
    restore-es.bat
  2. Verify the Elasticsearch database has been restored: "accepted":true

  3. Navigate to C:\Program Files and delete folder Elasticsearch 5.5.0.

Upgrading the UCP Advisor software

Upgrading to UCP Advisor version 2.9.0 requires installing the following two components on the UCP Advisor controller VM:
  • UCP Advisor software
  • UCP Advisor API
NoteThe UCPAdvisorOnBoardSheet.csv onboarding template from versions prior to 2.8.0 cannot be used in this upgraded version. A new template file is available in the same location (C:\Program Files\Hitachi\UCP Advisor\Documentation).

Installing the UCP Advisor software

After upgrading the third-party software, perform the following procedures on the UCP Advisor controller VM to install the UCP Advisor software.

Before you begin

  1. Preserve the Elasticsearch database and uninstall Elasticsearch 5.5.0.
  2. Upgrade Java 8 to OpenJDK 11.0.2.
  3. Install Elasticsearch 6.2.2.
  4. Restore the Elasticsearch database.

Procedure

  1. Browse to the location of the UCP Advisor installation media:

    F:\Software\UCPAdvisor29
  2. Double-click the UCPAdvisor-Master-setup-02.9.0-(x64).exe file.

  3. Click Install.UCP advisor sw install wizard

  4. Click Next.UCP advisor install wizard next

  5. Accept the terms of the license, and then click Next.

  6. Click Install, and then wait for the installation to complete.

Installing the UCP Advisor REST API

Install the UCP Advisor REST API on the UCP Advisor controller VM.

  1. Browse to C:\Program Files\Hitachi\UCP Advisor\RestAPI.

  2. Double-click setup.exe to launch the install wizard.

  3. Click Next.UCP Advisor sw install wizard next

  4. Accept the terms of the license, and then click Next.

  5. Enter the customer information, and then click Next. UCP advisor sw rest API install wizard customer info

  6. Set the destination folder for the installation. Click Next to accept the default folder, or click Change to specify a different folder.UCP Advisor rest API install wizard dest info

  7. Click Install to begin the installation.UPS Advisor rest api install ready

  8. Click Finish.

  9. Restart the VM.

Registering the UCP Advisor plugin

Register the UCP Advisor plugin using UCP Advisor Toolbox.

Procedure

  1. From the UCP Advisor controller VM desktop, launch the UCP Advisor Toolbox.

  2. In the vCenter plugin pane, enter the following information:

    • VCSA Address: The vCenter Server Appliance Fully Qualified Domain Name (FQDN). Enter it exactly as it appears on the Host and Clusters tab in the Navigator pane of the vSphere Web Client.
    • Root Password: The VCSA password for the root user.
    • SSO Username: The VCSA single sign-on (SSO) root username.
    • SSO Password: The VCSA SSO password for the root user.
  3. Click Submit.

  4. At the following UCP Advisor ToolBox prompt, click Yes:

    Existing UCP plugin version <2.8.0.485> from UCP Server <address> is registered with <vCenter address>. Do you want to upgrade the plugin to <2.9.0.148>?
  5. Click OK.

  6. Close the UCP Advisor ToolBox.

Next steps

Verify that UCP advisor plugin registered correctly.

  1. Log on to the vSphere Web Client: https://<vSphereVM_address>/vsphere-client/?csp.
  2. Verify that the Hitachi UCP Advisor for VMware vCenter plug-in is present on the home page.
    NoteIt will take some time to discover the devices and onboard them again.
  3. In the Navigator pane, click Administration > Client Plugins, and verify that latest version is installed. For current versions, see the UCP Advisor Release Notes.
NoteIf the plugin does not appear, restart the VCSA web client. For information on restarting the VCSA web client, see: https://docs.vmware.com/en/VMware-vSphere/6.7/com.vmware.vsphere.vcsa.doc/GUID-65843185-BD8E-49A8-904C-9419A907CF1C.html.

Updating UCP Advisor configuration files

If the customer modified the UCP Advisor XML configuration files after installing the previous version of the UCP Advisor software, you need to update the upgraded XML configuration files using the XML files you backed up when preparing for this upgrade. Skip this procedure if the customer did not modify the configuration files.

Before you begin

Locate the XML configuration files you backed up when preparing for this upgrade. They should be located in C:\temp\AdvisorConfigBackupFor2_8.

Procedure

  1. In an administrator command window, compare each of the old and new XML configuration files for differences between them.

  2. Manually update the new files with the values found in the old files. The new files are located in C:\Program Files\Hitachi\UCP Advisor\.

    For example, a comparison of the old and new versions of the App.config file shows a difference in the values for the CommunicationType key. The old value is HTTPS but the new value is HTTP. You would update this key in the new App.config file with the value from the old App.config file.

    fc "C:\Program Files\Hitachi\UCP Advisor\HiLogInsightConverged\SyslogAgent\App.config"
    C:\temp\AdvisorConfigBackupFor2_8\HiLogInsightConverged\SyslogAgent\App.config
    ***** C:\PROGRAM FILES\HITACHI\UCP ADVISOR\HILOGINSIGHTCONVERGED\SYSLOGAGENT\App.config
    <!-- CommunicationType possible value: HTTP or HTTPS -->
    <add key="CommunicationType" value="HTTP"/>
    <add key="httpsPort" value="2054"/>
    ***** C:\TEMP\ADVISORCONFIGBACKUPFOR2_8\HILOGINSIGHTCONVERGED\SYSLOGAGENT\APP.CONFIG
    <!-- CommunicationType possible value: HTTP or HTTPS -->
    <add key="CommunicationType" value="HTTPS"/>
    <add key="httpsPort" value="2054"/>
    *****
    

Verifying the status of the UCP Advisor plugin, services, and devices

Verify the status of the UCP Advisor plugin, services, and all onboarded devices. Verify that the version numbers match those listed in the Release Notes.

Before you begin

Access the Release Notes for the upgrade version: https://knowledge.hds.com/Documents/Converged/UCP_Advisor.

Procedure

  1. From the UCP Advisor controller VM, open Control Panel > Programs and Features.

  2. In the Version column, verify that the version of the UCP Advisor plugin matches the version listed in the Release Notes.

  3. Open Administrative Tools Services, and then verify that the following services are running. If a service is not running, start it.

    • Elasticsearch 6.2.2
    • Hitachi Syslog Agent for vRLI
    • Hitachi Baremetal
    • Hitachi UCP Advisor Ghost
    • Hitachi Hal
    • Hitachi UCP Advisor Hi-Track Web Service
    • Hitachi UCP Advisor Service Registry
    • Hitachi UCP Advisor Services
    • Hitachi UCP Advisor Sidecar
    • Hitachi UCP VI Service
    • Hitachi vRO Converge Web Service
    • Hitachi Watchdog Service
  4. Log on to the vSphere Web Client.

  5. In the Navigator pane, click Administration > Client Plug-Ins.

  6. In the Version column, verify that the version of the UCP Advisor plugin matches the version listed in the Release Notes.

  7. Navigate to the UCP Advisor plugin. The UCP Advisor Dashboard shows all onboarded devices.

    The status for all onboarded devices may appear as Unreachable. Wait for the database to refresh, possibly for 30 to 45 minutes. The status will refresh to Normal, indicating that the software upgrade is complete.

Upgrading management ESXi hosts

Make sure that the ESXi hosts are updated to the version supported by UCP Advisor.

Procedure

  1. Verify that the management hosts are clustered, so they can failover the management VM for the dedicated management host.

  2. Verify that the compute host in the cluster has sufficient spare capacity to handle at least one node being down. If not, verify that the customer can accept the workload.

  3. Follow the VMware standard practice to upgrade the host. For supported versions, see the UCP Advisor Release Notes here: https://knowledge.hds.com/Documents/Converged/UCP_Advisor.

    Recommended patch: https://kb.vmware.com/s/article/2151102.

    For details on using the update manager for patch updates, see: https://docs.vmware.com/en/VMware-vSphere/6.7/com.vmware.vcenter.upgrade.doc/GUID-043EF6BD-78F7-412F-837F-CBDF844F850C.html.

Performing a final validation and cleanup

Verify that each device on the list you created when preparing for the upgrade is up and running.

Repeat the visual validation of vSphere for compliance and other issues to ensure consistency of the UCP Advisor system before and after the upgrade. Remove the VM snapshots taken before starting the upgrade.

Troubleshooting

Tips are provided for troubleshooting the upgrade process.

Using a snapshot to recover a previous VM version

You can restore a previous VM version from the snapshot taken during the upgrade process.

NoteRestoring snapshots can cause the host to go out of synchronization.

Procedure

  1. Power off all of the VMs according to the order specified in Taking VM snapshots.

  2. Restore all of the VMs based on the snapshot procedure described in Taking VM snapshots.

  3. Power the VMs back on in the following order:

    1. vSphere VM (wait approximately 15 minutes after powering on for all services to start up)

    2. vRealize Log Insight (VRLI , if installed)

    3. vRealize Orchestrator (VRO, if installed)

    4. Hitachi Automation Director (HAD, if installed)

    5. UCP Advisor Gateway VM

    6. UCP Advisor controller VM (wait 5 minutes)

    7. Wait 1-2 minutes before starting the next VM.

Fixing host synchronization errors

When reverting to a VM snapshot, vCenter can generate an error indicating that the host cannot be synchronized. You can fix these synchronization errors using the following procedure.

  1. From the vSphere Web Client, right-click the host, then click Connection > Disconnect.

  2. Follow the prompts, and then wait for the current task to complete.

  3. Click Connection > Connect.

  4. Provide the appropriate credentials to reconnect.

    The synchronization error message should be resolved.

Troubleshooting switch backup and upgrade issues

Use the following guidelines and suggested steps to help resolve backup switch configuration and switch firmware upgrade issues.

  • Condition:

    An issue occurs when backing up switch configurations or upgrading switch firmware for Brocade G620 and Cisco 3048, 93180YC, or 93180LC switches when the SSH host keys have been changed.

    Following is an example of the type of error message that is output for this condition.

    Error - Failed to upgrade firmware. Failed to upgrade network switch firmware with firmware file /UCPFirmware/Firmware/FC/G620/v8.0.1. Failed to upgrade network switch firmware with firmware file /UCPFirmware/Firmware/FC/G620/v8.0.1. One or more parameters are not valid. The server is inaccessible or firmware path is invalid. Please make sure the server name/IP address and the firmware path are valid, the protocol and authentication are supported. It is also possible that the RSA host key could have been changed and please contact the System Administrator for adding the correct host key. at Hitachi.Storage.ExceptionHelper`2.ThrowNewException (L logger, Exception exception, T id, Object[] args)
  • What it Means:

    As a security precaution, the backing up of switch configurations or upgrading of switch firmware fails when the SSH host keys have been changed.

  • Corrective Action:

    Clear the list of trusted SSH hosts for switches where the backup configuration or upgrade fails.

For Brocade switches, issue the following:

admin> sshutil delknownhost -all This Command will delete all the known host keys. Please Confirm with Yes(Y,y), No(N,n) [N]: y All Known Host(s) deleted Successfully.

For Cisco System switches, issue the following:

R1-93180LC-EX-B# clear ssh hosts

R1-93180LC-EX-B#

Upgrading firmware

This appendix describes how to manually upgrade the DS120 and DS220 firmware.

Upgrading the DS120 or DS220 firmware

UCP Advisor 2.8 is currently only certified to run version 4.23.06, follow these steps to upgrade to 4.23.06 from 3.16.06 or 3.75.06 using Toolkit.

NoteThe Toolkit detects all server nodes via IPv6 link local addresses and flashes the BIOS/BMC/CPLD/HBA/NIC. It also backs up the BIOS and BMC settings and reapplies it to the server nodes after flashing to preserve settings.

Procedure

  1. Start a Remote Desktop session, then log on to the UCP Advisor controller VM.AppendixA_RemoteDesktopConnection

  2. If not already mounted, mount the UCP MediaBundle ISO as F.

  3. Open a new administrator command window, and then copy the UCPToolKit software to a temp directory.

    md "c:\Temp\UCPToolKit"
    cd "c:\Temp\UCPToolKit"
    xcopy /cedy F:\Tool\UCPToolKit
    
  4. Navigate to the C:\Temp\UCPToolKit\Requirements folder.Navigate to UCPToolKit requirements folder.

  5. Install vcredist_x86.AppendixA Install vcredist x86

  6. Navigate to C:\Temp\UCPToolKit\WinPython-64bit-3.6.3.0Zero.

  7. Right-click on WinPython Command Prompt and select Run as administrator.AppendixA_Right-click on WinPython Command Prompt and select Run as administrator

  8. On the command line, enter python firmware.py.

  9. Enter the number of nodes in the rack.

  10. Enter y for flashing existing nodes in a system.

  11. Enter the username of the BMC.

  12. Enter the password of the BMC.AppendixA Enter the password of the BMCs

  13. Verify you have entered the correct credentials and enter y.

  14. When prompted which appliance to install, enter 2.0.

  15. NoteUse this option only; this upgrade does not support any other option.
    AppendixA_When asked which appliance to install type 2.0
  16. Press Enter to continue.

  17. After the detection of nodes is complete, press Enter to continue. Only the powered off nodes will be updated.AppendixA_Once detection is completed, hit enter to continue

  18. Verify the BIOS, BMC, CPLD, HBA, or NIC firmwares are updated on the servers.AppendixA This will update the BIOS BMC CPLD HBA NIC firmware’s on the servers.

  19. Log on to UCP Advisor and refresh all servers to make sure the firmware version is 4.23.06.AppendixA_Now login to UCP Advisor and refresh all servers to make sure the firmware version is shown 4.23.06

  20. Delete the UCPToolKit folder under c:\Temp after all the blades are successfully upgraded.