Installing Cisco UCS Director on VMware vSphere

for VMware vSphere


Note

The appliance and boot-up logs are located in the /var/log/ucsd directory.

  • install.log contains the one time appliance installation logs.

  • bootup.log contains the appliance boot-up sequence information, such as startup messages for the database and infrastructure services.

Default Root and Shelladmin Passwords

During installation, Cisco UCS Director uses default passwords for the following accounts:

  • Root user for the CentOS operating system of the Cisco UCS Director VM. The default password is cisco123.

  • Shelladmin user for the Cisco UCS Director Shell menu. The default password is changeme.

Once the installation is completed ,the first time you log in to Cisco UCS Director, you are prompted to reset the default root and Shelladmin passwords.

The new root and Shelladmin password must meet the password requirements. It cannot be a dictionary word or be all lowercase.

Prerequisites for VMware vSphere

Before you install Cisco UCS Director for VMware vSphere, complete the following steps:

  • Install VMware vSphere or vCenter.

  • Configure a VMware vSphere or vCenter user account with system administrator privileges for Cisco UCS Director.

    You need administrator privileges to connect to and install Cisco UCS Director on VMware vCenter. Cisco UCS Director requires a user account with system administrator privileges to discover, manage and automate VMware vCenter configuration from Cisco UCS Director. These operations include creating, deleting and modifying VMs, ESXi hosts and clusters, datastores and datastore clusters, standard and DV switches, and virtual network port groups.

  • Download the Cisco UCS Director software from the Download Software area on Cisco.com.

  • Extract the Cisco UCS Director OVF file from the digitally signed zip file to your local disk. See Digitally Signed Images.

Minimum System Requirements for a Single-Node Setup

The following tables detail the minimum resource requirements for a single-node setup of . Cisco recommends a single-node setup for installations of up to 5000 VMs.

For optimal performance, the entire memory and CPU allocations specified in the table below should be reserved. Failure to follow these specifications could affect performance. For example, 4 vCPU cores with 3000 MHz and 16G of memory must be reserved for the VM.

The minimum memory required for the inframgr service is automatically set during deployment. To enable the inframgr service to use more than the minimum required memory, edit the inframgr.env file available in the following location:

/opt/infra/bin/

In this file, update the MEMORY_MAX parameter to the value you want. To activate the changes, restart the inframgr service.

The default memory settings are MEMORY_MIN=8192m and MEMORY_MAX=8192m.

For information about minimum system requirements for a multi-node setup, see the Cisco UCS Director Multi-Node Installation and Configuration Guide.

Table 1. Minimum system requirements for a single-node installation (up to 5,000 VMs)
Element Minimum Supported Requirement

vCPU

4

Allocated Memory

16 GB

Reserved Memory

16 GB

Disk Space

100 GB

Disk Write I/O Bandwidth

4 MBps

Disk Read I/O Bandwidth

4 MBps

Memory Allocated for inframgr

8 GB

Restart the database and all services after making these changes to the /etc/my.cnf.

Up to 2,000 VMs

If you plan to manage up to 2,000 VMs, the environment must meet at least the minimum system requirements in the following table.

Table 2. Minimum System Requirements for up to 2,000 VMs
Element Minimum Supported Requirement

vCPU

4

Memory

16 GB

Primary Disk (Hard Disk 1)

100 GB

Secondary Disk (Hard Disk 2)

100 GB

Disk Read I/O Bandwidth

4 MBps

Disk Write I/O Bandwidth

4 MBps

Up to 5,000 VMs

If you plan to manage no more than 5,000 VMs, the environment must meet at least the minimum system requirements and recommended configurations in the following tables.

Table 3. Minimum System Requirements for up to 5,000 VMs
Element Minimum Supported Requirement

vCPU

8

Memory

20 GB

Primary Disk (Hard Disk 1)

100 GB

Secondary Disk (Hard Disk 2)

100 GB

Disk Write I/O Bandwidth

4 MBps

Disk Read I/O Bandwidth

4 MBps

You must also edit the MEMORY_MIN and MEMORY_MAX settings in /opt/infra/bin/inframgr.env as follows:

MEMORY_MIN=8192m

MEMORY_MAX=8192m

Restart the inframgr service after making the memory size changes.

Edit the following parameters in the /etc/my.cnf file.

Table 4. Minimum Database Configuration
Element Minimum Supported Configuration

thread_cache_size

100

max_connections

1000

innodb_lock_wait_timeout

100

query_cache_size

128 MB

innodb_buffer_pool_size

2048 MB

max_connect_errors

10000

connect_timeout

20

innodb_read_io_threads

64

innodb_write_io_threads

64

Restart the database and all services after making these changes to the /etc/my.cnf.

Installing on VMware vSphere


Note

We recommend that you use VMware vCenter for OVF deployment. VMware vCenter versions 5.x and above are supported. OVF deployment wizards support only IPv4 addresses. If you require IPv6, deploy the OVF with IPv4 addresses and then use the ShellAdmin to configure IPv6 addresses.

Before you begin

You need administrator privileges to connect to VMware vCenter. Cisco UCS Director requires a user account with system administrator privileges to discover, manage and automate VMware vCenter configuration from Cisco UCS Director. These operations include creating, deleting and modifying VMs, ESXi hosts and clusters, datastores and datastore clusters, standard and DV switches, and virtual network port groups.


Note

If you do not want to use DHCP, you need the following information: IPv4 address, subnet mask, and default gateway.

Procedure

Step 1

In the Navigation pane, choose the Data Center where you want to deploy Cisco UCS Director.

Step 2

Choose Datacenter > Deploy OVF Template.

Step 3

In the Source pane, do one of the following to choose your OVA source location:

  • Choose files , navigate to the location where you downloaded the OVF, choose the OVA file, and click Open.
  • Replace FQDN (Fully Qualified Domain Name) with the path to the URL on your local area network where the OVA file is stored, including the IP address or domain name, and click Next.
Step 4

In the Name and Location pane, do the following:

  1. In the Name field, edit the default VM name.

  2. From the Inventory Location area, choose the inventory location where is being deployed, and click Next.

    Note 

    If you chose a Data Center in Step 2, option b might not be available.

  3. Click Next.

Step 5

In the Compute Resource pane, choose the required host, and click Next.

Step 6

TheReview Details pane, will display template details ,verify and click Next.

Step 7

On the Select Storage , choose the storage location for the VM.

Step 8

In the Disk Format pane, choose one of the following options and click Next:

  • Thick Provisioned (Lazy Zeroed) format—To allocate storage immediately in thick format. This is the recommended format. All performance data is verified with this format.
  • Thick Provisioned (Eager Zeroed) format—To allocate storage in thick format. It might take longer to create disks using this option.
  • Thin Provisioned format—To allocate storage on demand as data is written to disk.
    Important 
    We recommend that you do not choose the Thin Provisioned format.
Step 9

In the NetworkMapping pane, choose the network and click Next
Step 10

In the Properties pane, enter the following information and click Next:

  • Management IP Address—The management IP address to be used for eth0. If your network uses DHCP, leave the default value of 0.0.0.0.

  • Management IP Subnet Mask—The management IP subnet mask to be used for eth0. If your network uses DHCP, leave the default value of 0.0.0.0.

  • Gateway IP Address—The Gateway IP Address to be used for eth0. If your network uses DHCP, leave the default value of 0.0.0.0.

  • Ucsd Root Password

  • ShellAdmin Password

Note 

While deploying the OVA in a Customise template window,if no default passwords are populated for root and shelladmin, please enter the default password cisco123 for root and changeme for shelladmin.

Step 11

In the Ready to Complete pane, do the following:

  1. Verify the options that you chose in the previous panes.

  2. Click Finishto start the deployment process.

Step 12

After the appliance has booted up, copy and paste the management IP address (from the IP address that is shown) into a supported web browser to access the Login page.

Step 13

On the Login page, enter admin as the username and admin for the login password.

Note 

We recommend that you change the default admin password after this initial login.
Step 14

Choose Administration > License.

Step 15

On the License page, click License Keys.

Step 16

Click Manage Personalities.

Step 17

On the Personality Configuration screen, check the required personalities.

You can check either UCSD or Big Data or both personalities if required.

Step 18

Click Submit.

For information about upgrading from Cisco UCSD 6.7.4.x to Cisco UCSD 6.8 system follow the migration process under the Cisco UCS Director Upgrade Guide, Release 6.8.

Reserving System Resources

For optimal performance, we recommend reserving extra system resources for beyond the minimum system requirements listed in Minimum System Requirements for a Single-Node Setup.


Note

For more information about how to reserve system resources, see the VMWare documentation.

Procedure

Step 1

Log in to VMware vCenter.

Step 2

Choose the VM for .

Step 3

Shut down the VM.

Step 4

In VMware vCenter, click the Resource Allocation tab to view the current resource allocations, and click Edit.

Step 5

In the Virtual Machine Properties pane, edit resource allocations by choosing a resource and entering the new values.

Step 6

Verify that the new resource allocations have been made.