cisco UCS Director Shell User Manual

cisco UCS Director Shell

Troubleshooting

This chapter contains the following sections:

• Backing up the Monitoring Database in a Multi-Node Setup,
• Pinging the Hostname and IP Address,
• Viewing Tail Inframgr Logs,
• Cleaning Up Patch Files,
• Collecting Logs from a Node,
• Collecting Diagnostics,
• Using Diagnostics Information,
• Troubleshooting VMware Console Display Issues,
• Enabling HTTP Access
• Resetting MariaDB User Password in a Multi-Node Setup,
• Resetting MariaDB User Password in a Standalone Setup,
• Generating Device ID

Backing up the Monitoring Database in a Multi-Node Setup

• Problem —You are unable to back up the monitoring database in a multi-node setup.
• Recommended Solution— Edit the dbMonitoringBackupRestore.sh script

1. Step 1 Edit the /opt/infra/dbMonitoringBackupRestore.sh script using vi.
2. Step 2 Remove the CHARGEBACK_HISTORY_ENTRY table name from the script.

Pinging the Hostname and IP Address

You can ping a hostname or IP address to test your connectivity by choosing the Ping Hostname/IP address option.

1. Step 1 From the Cisco UCS Director Shell menu, choose the Ping Hostname/IP address option and press Enter

2. Step 2 Enter your IP address and press Enter.
   The following information is displayed:

3. Step 3 Press Enter to exit out of the operation

Viewing Tail Inframgr Logs

This Shell lets enables you to see inframgr (Infrastructure Manager) log data, which are generated behind the scenes by use of the Unix tail command. When you are debugging, you can trace problems by using this log data. You use the Tail Inframgr Logs option to immediately tail the most recent inframgr logs. The results are displayed on your screen directly after you select this option.

1. Step 1 From the Cisco UCS Director Shell menu, choose the Tail Inframgr Logs option and press Enter.
   Following are a few sample lines, typical of the results displayed immediately after use of the Tail Inframgr Logs option

2. Step 2 To exit from the log file display, type Ctrl+C, then press Enter.

Cleaning Up Patch Files

1. Step 1 From the Cisco UCS Director Shell menu, choose the Clean-up Patch Files option and press the Enter key.
   The following information is displayed:
   Do you want to delete an old patch file/directory [y/n]?

2. Step 2 Enter y and press Enter to delete the patch files.
   The following information is displayed

3. Step 3 Choose the required option to delete the patch file or directory and press Enter.
   The following information is displayed:
   Select an option to delete a patch file/directory: 4
   Are you sure you want to delete: cucsd_patch_6_6_0_0_66480 [y/n]?

4. Step 4 If you are prompted to confirm that you want to delete the patch file or directory, enter y and then press Enter.
   The following information is displayed
   Directory Deleted
   Press return to continue…

5. Step 5 Press the Enter key to return to the main menu

Collecting Logs from a Node

The Collect Logs from a Node option lets you collect logs from the local node or from a remote node.

1. Step 1 From the Cisco UCS Director Shell menu, choose Collect Logs from a Node Status and press Enter.
   The following list of services appears

2. Step 2 Enter the logs collection option and press Enter.

3. If you choose to collect logs from the current node, a response similar to the following appears
   Note
   To collect logs from another node, the best practice is to return to the Shell menu, select the Collect Logs from a Node option again, and choose the Remote Node option

4. If you choose to collect logs from a remote node, a response similar to the following appears:
   Please enter the remote server IP/Hostname from where we collect logs:

5. Follow the onscreen instructions to provide the address of the remote log, establish a secure connection, and provide the required login credentials for that remote node

Collecting Diagnostics

The Collect Diagnostics option helps to collect logs from a Multi-Node setup and a Standalone setup for debugging purposes

1. Step 1 From the Cisco UCS Director Shell menu, choose Collect Diagnostics. The following information is displayed
   Enter option number [0 1 2 3 4 5 6]:
   NOTE
   In a multi-node setup, only Collect essential diagnostics option is supported in inventory and monitoring nodes

2. Step 2 If you choose Collect essential diagnostics option, a response similar to the following appears:

3. Step 3 If you choose Collect basic diagnostics option, a response similar to the following appears:

4. Step 4 If you choose Collect full diagnostics option, a response similar to the following appears

5. Step 5 Enter the JDK path and press Enter. The following information is displayed

6. Step 6 If you choose Collect inframgr thread dump option, a response similar to the following appears:

7. Step 7 Enter the JDK path and press Enter. The following information is displayed

8. Step 8 If you choose Collect inframgr heap dump option, a response similar to the following appears

9. Step 9 Enter the JDK path and press Enter. The following information is displayed

10. Step 10 If you choose Display Disk Speed option, a response similar to the following appears
    Type in option number and press : 6
    Results of the disk speed check are more accurate if UCS Director services and database services are not running. If these services are running when you initiate the disk speed check, it could affect the disk I/O bandwidth for these services and impact system performance. We recommend you stop these services using the shelladmin option while running this utility. Do you want to continue [y/n]? :
    Note
    Display Disk Speed option allows you to check the read/write speed of a database disk in the Cisco UCS Director appliance. In a dual-node setup, Display Disk Speed option is not applicable for a primary node, as the primary node does not run the database services.

11. Step 11 Enter y and press Enter. The following information is displayed.:

Using Diagnostics Information

User or TAC engineer can collect the basic diagnostics data using Collect basic diagnostics option in the shelladmin while reporting any issue. The diagnostics bundle contains the following diagnostics data that is used for troubleshooting the reported issues

• Summary file —Contains important and high level summary.
• Diag file —Contains information such as version history with timestamp, average CPU utilization, infra services status, database status, and database size.
• SummaryReport file —Contains summary report.
• DiagOutput file —Contains detailed report.
• UcsdExceptions file —Contains all exceptions found in the inframgr/logfile.txt.* and number of occurrences of each exception.
• infra-env Directory —Contains the infra services configuration (.env) files.
• commands Directory —Contains the output of various system commands.
• var-log-ucsd zip file —Contains the log files such as install.log, bootup.log, and services.lo

Troubleshooting VMware Console Display Issues

• Problem— The VMware console does not display after an abrupt shutdown of the Cisco UCS Director VM from VMware vCenter.
• Possible Cause —Occasionally after Cisco UCS Director VM is powered on, the VMware console prompt\ gets stuck after the process restart and does not return to the shelladmin.
• Recommended Solution— After the VM is powered on, press Alt-F1 to refresh the VMware console.
  In the Cisco UCS Director VM prompt after the VM is powered on, press Alt-F1.
  The VMware console screen is refreshed

Enabling HTTP Access

By default, HTTPS access mode is enabled during initial OVF installation and Cisco UCS Director upgrade. When HTTP is enabled, you can log in to Cisco UCS Director, using both HTTP and HTTPS modes. When HTTPS is enabled, you can log in to the Cisco UCS Director only using HTTPS mode. Even when you try to log in to Cisco UCS Director using HTTP mode, you will be redirected to HTTPS user interface only

1. Step 1 From the Cisco UCS Director Shell menu, choose the Enable/Disable HTTP option and press Enter.
   The following information is displayed:
   HTTPS is currently enabled. Do you want to enable HTTP [y/n]? :

2. Step 2 Enter y and press Enter.
   The following information is displayed:
   Cisco UCS Director Services will be restarted to enable the HTTP configuration. Do you want to continue [y/n]?

3. Step 3 Enter y and press Enter. The Cisco services are restarted.

Resetting MariaDB User Password in a Multi-Node Setup

1. Step 1 From the Cisco UCS Director Shell menu, choose the Reset MariaDB User password option and press Enter.
   The following information is displayed:
   This utility will restart the services after changing MariaDB user password, do you want to continue? [y/n]:
   In a multi-node setup, ensure that the infra services are stopped in the primary and service nodes before executing the Reset MariaDB User password option in DB nodes.

2. Step 2 Enter y and press Enter.
   The following information is displayed:

3. Step 3 Enter y and press Enter
   The following information is displayed
   Current Password (Type in current password or press enter key to use password from the existing credentials file):
   This option is applicable only for the primary and service nodes in a multi- node setup.

4. Step 4 Enter y and press Enter.
   The following information is displayed:
   Do you want to generate random password for MariaDB ‘admin’ user? [y/n]:

5. Step 5 Enter n and press Enter.
   The following information is displayed:
   Specify the new password for MariaDB ‘admin’ user:

6. Step 6 Enter a new MariaDB admin password and press Enter.
   Note Special characters such as *, \, ‘, and $ are not allowed for MariaDB admin user passwords

7. Step 7 Enter your new MariaDB admin password and press Enter.
   The following information is displayed
   Note If a BMA appliance is associated with a Cisco UCS Director, the dbkeys and dbcreds files are copied to a specific location in the BMA appliance to establish successful connectivity to the Cisco UCS Director. After resetting the MariaDB user password, you should restart the BMA services either from the Cisco UCS Director user interface or from the BMA appliance
   Note In a multi-node set up, if you want to reset the MariaDB user password, you should execute the Reset MariaDB User password option in all the nodes in the following sequence inventory, monitoring, primary, and service nodes.

Resetting MariaDB User Password in a Standalone Setup

1. Step 1 From the Cisco UCS Director Shell menu, choose the Reset MariaDB User password option and press Enter.
   The following information is displayed:
   This utility will restart the services after changing MariaDB user password, do you want to continue? [y/n]:

2. Step 2 Enter y and press Enter.
   The following information is displayed:
   Stopping the infra services…
   The infra services are stopped.
   Do you want to change the password for MariaDB ‘admin’ user? [y/n]:

3. Step 3 Enter y and press Enter.
   The following information is displayed:
   Do you want to generate random password for MariaDB ‘admin’ user? [y/n]:

4. Step 4 Enter y and press Enter.
   The following information is displayed:
   Generating Random Password……. done
   Do you want to change the password for MariaDB ‘root’ user? [y/n]:

5. Step 5 If you entered n, enter the new password for MariaDB admin user and press Enter.
   Note Special characters such as *, \, ‘, and $ are not allowed for MariaDB admin user passwords.
   The following information is displayed:
   Specify the new password for MariaDB ‘admin’ user:
   Confirm the new password for MariaDB ‘admin’ user:
   Password update takes few minutes. Please wait….. done

6. Step 6 Enter y and press Enter.
   The following information is displayed|
   Do you want to generate random password for MariaDB ‘root’ user? [y/n]:

7. Step 7 Enter y and press Enter.
   The following information is displayed:

8. Generating Random Password……. done

9. MariaDB user password is updated.

10. Checking if the database is running… yes.

11. Stopping the database…………….

12. The database is stopped.

13. Starting the database…

14. Checking if MariaDB database is running… .UP

15. The database is started.

16. Starting the infra services…

17. Step 8 If you entered n, enter the new password for MariaDB root user and press Enter.
    Note Special characters such as *, \, ‘, and $ are not allowed for root user passwords.
    The following information is displayed
    Note After resetting the MariaDB user password, you should restart the BMA services either from the Cisco UCS Director user interface or from the BMA appliance.

Generating Device ID

You can generate a device ID for a cloned Cisco UCS Director appliance by choosing Configure Network Interface option.

1. Step 1 From the Cisco UCS Director Shell menu, choose Configure Network Interface and press Enter.
   The following information is displayed:
   Cisco UCS Director’s VM UUID change detected. It is recommended to generate a new GUID for this UCS
   Director instance. Proceed [y/n]?
   Note
   This option is displayed only when a Cisco UCS Director is cloned. You must generate a new GUID. The GUID is used to claim a device in Cisco Intersight. For more information about how to claim a device, see the Cisco UCS Director Administration Guide

2. Step 2 Enter y to assign a new, unique, and unclaimed device ID to the cloned Cisco UCS Director, and press Enter. The following information is displayed
   Generation of Cisco UCS Director GUID is successful. After configuring the network interface, you must restart the Cisco UCS Director services for the updated network configuration to be used.
   Do you want to Configure DHCP/STATIC IP [D/S] ? :
   Note  Enter n only if you want Cisco Intersight to call the cloned Cisco UCS Director rather than the original Cisco UCS Director.
   Note To configure a network interface for the Cisco UCS Director appliance, see Configuring a Network Interface

Read User Manual Online (PDF format)

Read User Manual Online (PDF format)  >>

Download This Manual (PDF format)

Download this manual  >>