The documentation set for this product strives to use bias-free language. For the purposes of this documentation set, bias-free is defined as language that does not imply discrimination based on age, disability, gender, racial identity, ethnic identity, sexual orientation, socioeconomic status, and intersectionality. Exceptions may be present in the documentation due to language that is hardcoded in the user interfaces of the product software, language used based on RFP documentation, or language that is used by a referenced third-party product. Learn more about how Cisco is using Inclusive Language.
Cisco Modeling Labs has relatively modest storage requirements, with a 250GB capacity (or larger) Direct Attached Storage disk (DAS) recommended. RAID configurations are optional. When using a RAID configuration on the UCS C-Series server, the hardware based (MRAID module) version is the recommended method.
Storage Area Network (SAN) options are beyond the scope of this installation guide. SAN options are not supported for Cisco Modeling Labs bare metal deployments on Cisco UCS C-Series.
If the Cisco UCS C-Series server is being freshly deployed, there are some preliminary preparations that are necessary to prepare the hardware. These include configuring the server’s dedicated management interface (CIMC); verifying that the necessary Virtualization Technology features are enabled in the BIOS; and preparing the storage for the installation. The following steps are associated with the Cisco UCS C220 M4S platform running Version 2.06(6d) BIOS/CICM firmware. Refer to the applicable documentation if other server types or firmware levels are to be used and adjust the process accordingly.
With the CIMC interface configured, it is accessed to complete the machine preparation and to facilitate the software installation.
To map the Cisco Modeling Labs ISO disk image, complete the following steps:
Step 1 | With the Cisco
UCS server properly prepared for the Cisco Modeling Labs installation, the ISO
installation media must be virtually (remote) mounted to the target server. In
the CIMC interface, open a KVM Console to the server by clicking the associated
icon in the tool bar or the within the
Actions pane.
| ||
Step 2 | In the KVM
Console window, click
Virtual Media from the menu bar. From the drop-down, choose
the
Activate Virtual Devices. Acknowledge the
Unencrypted Virtual Media Session warning and click
Apply, as shown.
| ||
Step 3 | Click
Virtual Media from the menu bar again. In the expanded
drop-down list, choose the
Map CD/DVD… option. In the resultant Virtual Media – Map
CD/DVD dialog box, browse to and select the Cisco Modeling Labs ISO file. The
ISO image file will appear in the selected Drive/Image File field; click
Map Device to continue, as shown.
| ||
Step 4 | In the KVM Console window, click Macros from menu banner. In the drop-down list, choose Static Macros > Ctrl-Alt-Del to trigger a server reboot. | ||
Step 5 | During the reboot cycle, when the server setup screen is displayed, press the <F6> key. Choose the Cisco vKVM-Mapped vDVD option for the boot device. When complete, the server will boot the ISO disk image file. |
Live—Boots an non-persistent VIRL instance into the server. This method allow for changes to be made before the install is initiated. An Install System to HDD icon on the booted Ubuntu desktop initiates the software installation process.
Install—Skips the initial Ubuntu load and directly initiates the system installation process.
Memtest—Performs a test of the system DRAM.
Hd—Aborts the installation and restarts the system with the first hard disk.
Note | From this point onwards, installation actions should be performed using the directly attached console. Attempting to perform the installation process using the CIMC’s virtual KVM console may trigger a virtual screen resolution fault. |
To map the Cisco Modeling Labs ISO disk image, complete the following steps:
Step 1 | Choose
Install Option 1: live - boot VIRL for changes before
install. When the log in screen is displayed, log into the virtual
machine using username virl and password VIRL.
| ||||
Step 2 | On the
desktop, double-click
Install system
to HDD to begin the installation.
| ||||
Step 3 | In the Welcome window, choose the applicable language from the list and click Continue. | ||||
Step 4 | The
Preparing to install System window will request affirmation
that minimum requirements have been met. The recommended storage for Cisco
Modeling Labs is 250G or larger. Do not enable download updates or third-party
software while installing. Click
Continue.
| ||||
Step 5 | In the
Installation Type window, the Installation type
should be set to Erase disk and Install System.We recommend that you
enable the
Use LVM with the new System installation option, to setup
Logical Volume Management. Click
Install Now.
| ||||
Step 6 | Confirm the disk partitioning information displayed and click
Continue.
| ||||
Step 7 | In the Where are you? window, enter the applicable time zone, and click Continue. | ||||
Step 8 | In the Keyboard Layout window, choose the applicable locale-specific keyboard layout to use by clicking the corresponding option from the Choose your keyboard layout list, and click Continue. | ||||
Step 9 | In the
Who are you? window, enter values for the fields exactly as
described here:
Click Continue to start the installation. This initiates the image transfer process and copies the files to the UCS Server. A status window indicates the progress of the installation. Once the file transfer has completes, the system detects the running hardware; applies necessary adaptations, and performs a post-install cleanup. | ||||
Step 10 | The
Installation Complete notification indicates a successful
installation. Click
Restart Now to effect a reboot using the installed image.
| ||||
Step 11 | If you choose Install Option 2: install - start the VIRL installer directly, this bypasses booting a transient instance of the server and immediately starts the installation. Steps 3 -10 are performed as described above. |
The Cisco Modeling Labs bare-metal install requires 5 network interfaces, named eth0, eth1, eth2, eth3, and eth4. The presence of these interfaces should be verified at this point. Following install options 1 (live) or 2 (install), the Cisco Modeling Labs server is re-booted from the local disk. On completion of the reboot, log back into the console and open an xterm session.
Edit the /etc/default/grub file: sudo nano /etc/default/grub
GRUB_CMDLINE_LINUX_DEFAULT=”” GRUB_CMDLINE_LINUX=””
GRUB_CMDLINE_LINUX_DEFAULT=”biosdevname=0” GRUB_CMDLINE_LINUX=”biosdevname=0”
Save the /etc/default/grub file and exit using [Ctrl-X; Yes; Enter]
Complete the update using the command: sudo update-grub
Reboot the server to effect the changes: sudo reboot now
On completion of the system restart, verify that the required number of Ethernet interfaces conforming to the ethN naming format are now available on the operating system. If not, this must be diagnosed and resolved before proceeding, or the interface-constrained installation steps performed.
In accordance with best practices and to account for a possible lack of DHCP services, it is recommended that the eth0 interface be configured with a static IP address, as follows:
Step 1 | Start the
virtual machine and log in using the username virl and the password VIRL.
| ||
Step 2 | Click the xterm icon to open a terminal window. | ||
Step 3 | Change to the network interfaces configuration directory: cd /etc/network | ||
Step 4 | Open the interfaces configuration file for editing: sudo nano interfaces | ||
Step 5 | Change the eth0 addressing method to static: iface eth0 inet static | ||
Step 6 | Provide the static IP address: address n.n.n.n | ||
Step 7 | Provide the static IP address netmask: netmask mmm.mmm.mmm.mmm | ||
Step 8 | Provide the
default IP gateway address:
gateway
g.g.g.g
| ||
Step 9 | Provide valid reachable DNS name-server addresses: dns-nameservers a.a.a.a b.b.b.b | ||
Step 10 | Enter Ctrl-X to exit. | ||
Step 11 | Enter Y and Enter to confirm saving the interfaces file and exit. | ||
Step 12 | Enter sudo reboot now to reboot the virtual machine in preparation for the remaining installation steps. |
In a bare metal deployment, if the Cisco Modeling Labs server does not have the required 5 network interfaces, the missing interfaces require pseudo-interface (dummy) references. This is done by creating an alias for the missing OpenStack services IP address, and then creating a pseudo-interface for each of the missing interfaces.
The steps described here are for a server fitted with only two network interfaces (eth0 and eth1). Three pseudo-interfaces (dummy1, dummy2, and dummy3) must be configured to compensate for the missing interfaces. Adapt the number of pseudo-interfaces in accordance with the number required for your specific deployment. This section can be skipped if the server has the requisite five network interfaces.
Step 1 | From a console xterm session, edit the network configuration file: sudo nano /etc/network/interfaces | ||||
Step 2 | Add a new line
in the eth0 section and enter
up ip addr
add 172.16.10.250/24 dev eth0 to create a new alias for the missing
OpenStack services address.
For example:
iface eth0 inet dhcp dns-nameservers 8.8.8.8 8.8.4.4 up ip addr add 172.16.10.250/24 dev eth0
iface eth0 inet static address nnn.nnn.nnn.hhh netmask nnn.nnn.nnn.0 gateway nnn.nnn.nnn.g dns-nameservers 8.8.8.8 8.8.4.4 up ip addr add 172.16.10.250/24 dev eth0
| ||||
Step 3 | Open the configuration file for editing: sudo nano /etc/virl.ini | ||||
Step 4 | Change the hostname to ubuntu. This can be modified later during customization if desired. | ||||
Step 5 | Enter Ctrl-W and
search for 'l2_port:'.
| ||||
Step 6 | Enter Ctrl-W and search for 'l2_port2:'. In this example, since interface eth2 is missing, l2_port2: must be mapped to interface dummy1. Replace eth2 with dummy1. | ||||
Step 7 | Enter Ctrl-W and search for 'l3_port:'. In this example, since interface eth3 is missing, l3_port: must be mapped to interface dummy2. Replace eth3 with dummy2. | ||||
Step 8 | Enter Ctrl-W and search for 'internalnet_port:'. In this example, since interface eth4 is missing, internalnet_port: must be mapped to interface dummy3. Replace eth4 with dummy3. | ||||
Step 9 | Enter Ctrl-W and search for 'dummy_int'. Since dummy interfaces are required dummy_int must be set to True. | ||||
Step 10 | Enter Ctrl-X to exit nano. | ||||
Step 11 | Enter Y and Enter to confirm saving the configuration file and exit. | ||||
Step 12 | Enter sudo reboot now to reboot the virtual machine. | ||||
Step 13 | Once rebooted, log in again using username virl and password VIRL. | ||||
Step 14 | Click the xterm icon to open a terminal window. | ||||
Step 15 | Confirm that the
OpenStack services IP address is reachable:
ping -c 4
172.16.10.250
| ||||
Step 16 | Enter
nova
service-list to display the status of the Nova services.
Verify
that the status for each Nova service is enabled and that the state for each is
up.
| ||||
Step 17 | Enter
neutron
agent-list to display the status of the OpenStack Neutron agents.
Verify
that the status for the Metadata, DHCP, and L3 agents is :-).
|
Once the software has been installed on the server, changing the default video resolution will enable the Cisco Modeling Labs Desktop Manager GUI (Ubuntu Light Display Manager) to be accessible via the CIMC’s virtual KVM. This requires applying a shell script changing the default resolution to the lightdm configuration file.
Note | Changing the video resolution via the Desktop Manager’s GUI menu (Preferences > Monitor Settings) is ineffective, as it does not apply to the Login page, thus preventing remote logins. |
To manually set the video to a resolution supported by the CIMC’s virtual KVM, complete the following steps:
Step 1 | In the KVM Console window, click Macros on the menu bar. |
Step 2 | From the drop-down menu, choose the , followed by <Enter> to switch the vConsole to a command line interface (CLI). If necessary, login with virl/VIRL. |
Step 3 | Edit the lightdm.conf file: sudo nano /etc/lightdm/lightdm.conf |
Step 4 | Add the following line to the file: display-setup-script=/etc/lightdm/lightdm_cml.sh |
Step 5 | Save the file, and exit the editor: Ctrl-x; Yes; Enter |
Step 6 | Create a lightdm_cml.sh file: sudo nano /etc/lightdm/lightdm_cml.sh |
Step 7 | Add the following lines:
#!/bin/sh xrandr --output default --mode 1024x768 |
Step 8 | Save the file, and exit the editor: Ctrl-x; Yes; Enter |
Step 9 | Set the shell-script as executable by entering: sudo chmod +x /etc/lightdm/lightdm_cml.sh |
Step 10 | Reboot the machine using the command: sudo reboot now |
Step 1 | Once the virtual
machine completes the reboot cycle, establish a browser session to the Cisco
Modeling Labs server’s management interface (either the DHCP acquired address
noted earlier, or the static address added to the /etc/network/interfaces
file.)
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Step 2 | Click the
User
Workspace Management interface link. Login with the default
credentials (username= uwmadmin, password=password). The
User
Workspace Management Overview page is displayed.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Step 3 | From the options
on the left, expand the
CML Server
option and select
System
Configuration. Click
System to
set the system management details.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Step 4 | Click
Networks
to configure the other interfaces for external communications.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Step 5 | Click
VIRL
Services to configure the port numbers for VIRL services.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Step 6 | Click
Infrastructure to configure the other interfaces for
external communications.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Step 7 | Click
Resources
to configure the other interfaces for external communications to meet
integration requirements.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Step 8 | With all
configuration options set, click
Apply
Changes. A summary of the changes is presented, showing the previous
parameters settings and the new values being applied. Having confirmed that all
changes are correct, click
Apply
Changes at the bottom of the page.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Step 9 | Return to the Cisco Modeling Labs virtual machine console and open an xterm window. Initiate a system reboot with the sudo reboot now command. Alternatively, double-click on the 2. REBOOT icon on the desktop. When the system reboot has completed, return to the User Workspace Management interface to confirm the custom settings. |
Returning to the
User Workplace Management interface shows the server’s current licensing
status; the red banner indicates that there is no product licensing in place.
To license the Cisco Modeling Labs server, complete the following steps:
Step 1 | In the left
pane, click
Licenses.
The
Licenses page is displayed.
| ||||||||
Step 2 | In the Licenses page, click Register Licenses. | ||||||||
Step 3 | Record the
Host Name and
Mac Address for
license key registration.
Use this information when completing the Register Claim Certificates instructions in the eDelivery Order Notification email to request your license key for use with the Cisco Modeling Labs server.
You will receive your license key as an attachment via an email. | ||||||||
Step 4 | Open the attachment in a text editor and copy all of the contents. | ||||||||
Step 5 | Return to the
Register
Licenses
page and paste the details into the
Licenses
text area.
| ||||||||
Step 6 | Click
Register to
register the license key.
| ||||||||
Step 7 | Repeat Steps 4 – 6 for each license file received from the registration process. Verify that the Licenses page correctly reports the applied node count and expiration dates. | ||||||||
Step 8 | Click Log Out to exit the User Workspace Management interface. |