tripleo-docs/doc/source/basic_deployment/basic_deployment_gui.rst

Basic Deployment (GUI)

Access the GUI

Part of the Undercloud installation is also Tuskar-UI which you can use to drive the deployment.

Virtual

In the case of a virtual deployment, Tuskar-UI runs on the instack virtual machine on http://localhost/dashboard. Considering this and the fact that the virt host is a remote host machine, to access the UI in the browser, follow these steps:

1. On host machine create ssh tunnel from instack vm to virt host:

   ssh -g -N -L 8080:127.0.0.1:80 root@<undercloud_vm_ip>

2. On instack VM edit /etc/openstack-dashboard/local_settings and add virt host hostname to ALLOWED_HOSTS array

3. Restart Apache:

   systemctl restart httpd

4. Navigate to http://<virt_host_hostname>:8080/dashboard in the browser

When logging into the dashboard the default user and password are found in the /root/stackrc file on the instack virtual machine, OS_USERNAME and OS_PASSWORD.

After logging into the dashboard, make sure that the project is set to admin (if it is not, change the project to admin using the Project switcher at the top bar).

Overview Page

When you log into the GUI, you will land on the Overview page. This page contains all the information about the current state of the deployment. On the left side, there is a list of available deployment roles. On the right side, there is a deployment checklist which indicates whether all the prerequisites for the deployment have been satisfied.

Get Images

To perform a successful deployment, you will need the following images: discovery ramdisk, deployment ramdisk, and openstack-full image. To upload the images and load them into Glance, navigate to Provisioning Images and use the Create Image button to create the necessary images. You will need to following image files handy:

overcloud-full.vmlinuz
overcloud-full.initrd
overcloud-full.qcow2
deploy-ramdisk-ironic.kernel
deploy-ramdisk-ironic.initramfs
discovery-ramdisk.kernel
discovery-ramdisk.initramfs

To create the 'overcloud-full-vmlinuz' image, fill the Create Image form like so:

Name: overcloud-full-vmlinuz
Image Source: Image File
Image File: overcloud-full.vmlinuz
Format: AKI
Public: checked

To create the 'overcloud-full-initrd' image, fill the Create Image form like so:

Name: overcloud-full-initrd
Image Source: Image File
Image File: overcloud-full.initrd
Format: ARI
Public: checked

To create the 'overcloud-full' image, fill the Create Image form like so:

Name: overcloud-full
Image Source: Image File
Image File: overcloud-full.qcow2
Kernel: overcloud-full-vmlinuz
Ramdisk: overcloud-full-initrd
Format: QCOW2
Public: checked

To create the 'bm-deploy-kernel' image, fill the Create Image form like so:

Name: bm-deploy-kernel
Image Source: Image File
Image File: deploy-ramdisk-ironic.kernel
Format: AKI
Public: checked

To create the 'bm-deploy-ramdisk' image, fill the Create Image form like so:

Name: bm-deploy-ramdisk
Image Source: Image File
Image File: deploy-ramdisk-ironic.initramfs
Format: ARI
Public: checked

You will also need to copy the discovery images to HTTP BOOT directory on the undercloud node. Assuming you have these images handy in your home directory on the undercloud node, run the following commands to copy them to /httpboot:

sudo cp -f discovery-ramdisk.kernel /httpboot/discovery.kernel
sudo cp -f discovery-ramdisk.initramfs /httpboot/discovery.ramdisk

Register Nodes

To register nodes for your deployment, navigate to the Nodes page, and click the + sign to open the Register Nodes form. This form gives you a choice of manually adding node data using the Add link, or supplying a CSV file with node data using the Upload link. Choose one of these two options to register your nodes with Ironic. Make sure you provide, at minimum, the Power Management properties, the MAC address(es) for the nodes and the Deployment Images. The rest of the attributes are optional and in case you don't supply them, they will be obtained by running the introspection on the nodes. When you have finished providing the node data, click the Register Nodes button to register your nodes.

Note

When using a CSV file to upload node data, make sure the fields are in this order:

driver,address,username,password/ssh key,mac addresses,cpu architecture,number of CPUs,available memory,available storage

As stated above, hardware properties (cpu architecture, number of CPUs, available memory, available storage) are not mandatory, and in case they are not provided in the CSV file, they will be obtained via introspection.

Here are the contents of an example CSV file to register two virtual nodes:

pxe_ssh,192.168.122.1,root,"-----BEGIN RSA PRIVATE KEY-----MIIEowIBAAKCAQEAw6J6supEV40toLTiH6Taj8k6bI2CSJSK31spMfLIKzPuvzCV PGZdhKMx1o++u9TcFFh7U1caojg1Jj/XKdPcktGBQvAmiNa9nybmTjiOqq/b1svr W1Yn23WvkEBL7peFUZVAgJDvxcf42LtA72RdgzegFdrP0y4z6UJlJRnxAckxfa/o b05N3nrK2yteZQVuMBVB2P7QAgy62aIqJBacWrLplaZMJZZYQJ9ialXZXAMPIN3P 5l9emMOJjBbXv76G6B/Ik9J6Ryv8SFhQbLzWu1eXjT8g3d5KlD/lvh6fwN/HjYOb 6o5LvMD61vpOaR0B8Ta/+vu4R+GiLB+ArhS9WQIDAQABAoIBAAUWLGqKfMxp902+ ZkK8XpJugP5hj4mjkxjLnf6WeW3mI8cE1FyFrNjOBXW2txbbKf29bzdzhFXDcF5W Opnz0EBhAiNjax0TuEpzEHnoLo1xlR24n534V4D1RmNRyKejeOvuHYc6PYG++VFp TP7sdSH8SEVJMy4ifWcLOuYEDqglL1uSPQgO8HkmlvOvgI1LnLx9wjeNC1D7weZu Eh75GTGRLL8i0X0bLmaNZ1Fs3Ge2tLNS0hfu6epCiT3ZAQTBVyFbVmN6btnQ/BHM nDSZQ2JEDjQByCiBch9hTk/V2UNmn5dOUGPTwp2IP5Blpq2X7u1IoXQiAhI+zVcN 9mqbK6ECgYEA9hbQF7iEP4RhClNUVmQJd1zQjC2D5Vj0ik68MdgcT1QFrrCGaRPb eayCxyDoNyUAWGtqToTQ5v3b1dxwsJryMekHD0OL75fl1KbD0bRiawVG81QbyC3U I05Lr5LCdm80xdktC6caIkpoRF6e9xhAQduXDHZyQ6TdJtRHS6E3uPsCgYEAy4N6 xFml63vk2qFPyMkSfp82ey6YiGchXxZSDl/tDiDDHgDVRtYi5+7iCNVrjkwtUXI3 zK+G+m74AIx1C9ZSW81y5ymxKLGz1+OUy3Vtp0Zf5R1/Q+l9I4sl1dkB2wJcb2Ls 2A3yl9NEt5M3bHZUQk4ttmhfqOFiSmNd/uFbersCgYEAvlAYMDAPfnum/HBDKeiF dZz+31mXxjeMLqYDXtzNz/+fwWBdIkgsFKX4IX1ueK8R3E990Clg0TMD3xlywPGj WjvnrMNFJk6nfFRX3gaNkkxreWTTc3UVuRQk7iwmXadU/akd8AQT7u7yQeWXNGq5 zvS+lPHJHk0ShqPmWzPbvx8CgYBAiB9slXSsN+v4e4AeDcwkhH21D7BkSDdnvF8m mbpEaZUVNXRrcsk8vB3GaU4in/sawVn6OIpXbMqM+fy/VSVmYL4XmLvJSJfbVBnB binoCcOsle7d7PK2S5AiwB37gUMoOrkZRUrwY5h9kVvYs6jCIaITHgN/PIB7UAjl IjZsswKBgC9AgnXvw4M1bcS1SK1WdJXACrmfX5tGMLzCEVJgmJtiWobvpNsfcZ/Q EanIrYxnJ1zWZstefEuLWGzja+xwL/rsnTl77DPuvZRr/QxXMRaPFn5VTnH5kh0y 9FlihAmgG1n2k3CCFNtdxAKBqPGLz2wUuRDHYhF4WKYuvghIpQA7 -----END RSA PRIVATE KEY-----",00:d1:2c:a2:ed:58 pxe_ssh,192.168.122.1,root,"-----BEGIN RSA PRIVATE KEY-----MIIEowIBAAKCAQEAw6J6supEV40toLTiH6Taj8k6bI2CSJSK31spMfLIKzPuvzCV PGZdhKMx1o++u9TcFFh7U1caojg1Jj/XKdPcktGBQvAmiNa9nybmTjiOqq/b1svr W1Yn23WvkEBL7peFUZVAgJDvxcf42LtA72RdgzegFdrP0y4z6UJlJRnxAckxfa/o b05N3nrK2yteZQVuMBVB2P7QAgy62aIqJBacWrLplaZMJZZYQJ9ialXZXAMPIN3P 5l9emMOJjBbXv76G6B/Ik9J6Ryv8SFhQbLzWu1eXjT8g3d5KlD/lvh6fwN/HjYOb 6o5LvMD61vpOaR0B8Ta/+vu4R+GiLB+ArhS9WQIDAQABAoIBAAUWLGqKfMxp902+ ZkK8XpJugP5hj4mjkxjLnf6WeW3mI8cE1FyFrNjOBXW2txbbKf29bzdzhFXDcF5W Opnz0EBhAiNjax0TuEpzEHnoLo1xlR24n534V4D1RmNRyKejeOvuHYc6PYG++VFp TP7sdSH8SEVJMy4ifWcLOuYEDqglL1uSPQgO8HkmlvOvgI1LnLx9wjeNC1D7weZu Eh75GTGRLL8i0X0bLmaNZ1Fs3Ge2tLNS0hfu6epCiT3ZAQTBVyFbVmN6btnQ/BHM nDSZQ2JEDjQByCiBch9hTk/V2UNmn5dOUGPTwp2IP5Blpq2X7u1IoXQiAhI+zVcN 9mqbK6ECgYEA9hbQF7iEP4RhClNUVmQJd1zQjC2D5Vj0ik68MdgcT1QFrrCGaRPb eayCxyDoNyUAWGtqToTQ5v3b1dxwsJryMekHD0OL75fl1KbD0bRiawVG81QbyC3U I05Lr5LCdm80xdktC6caIkpoRF6e9xhAQduXDHZyQ6TdJtRHS6E3uPsCgYEAy4N6 xFml63vk2qFPyMkSfp82ey6YiGchXxZSDl/tDiDDHgDVRtYi5+7iCNVrjkwtUXI3 zK+G+m74AIx1C9ZSW81y5ymxKLGz1+OUy3Vtp0Zf5R1/Q+l9I4sl1dkB2wJcb2Ls 2A3yl9NEt5M3bHZUQk4ttmhfqOFiSmNd/uFbersCgYEAvlAYMDAPfnum/HBDKeiF dZz+31mXxjeMLqYDXtzNz/+fwWBdIkgsFKX4IX1ueK8R3E990Clg0TMD3xlywPGj WjvnrMNFJk6nfFRX3gaNkkxreWTTc3UVuRQk7iwmXadU/akd8AQT7u7yQeWXNGq5 zvS+lPHJHk0ShqPmWzPbvx8CgYBAiB9slXSsN+v4e4AeDcwkhH21D7BkSDdnvF8m mbpEaZUVNXRrcsk8vB3GaU4in/sawVn6OIpXbMqM+fy/VSVmYL4XmLvJSJfbVBnB binoCcOsle7d7PK2S5AiwB37gUMoOrkZRUrwY5h9kVvYs6jCIaITHgN/PIB7UAjl IjZsswKBgC9AgnXvw4M1bcS1SK1WdJXACrmfX5tGMLzCEVJgmJtiWobvpNsfcZ/Q EanIrYxnJ1zWZstefEuLWGzja+xwL/rsnTl77DPuvZRr/QxXMRaPFn5VTnH5kh0y 9FlihAmgG1n2k3CCFNtdxAKBqPGLz2wUuRDHYhF4WKYuvghIpQA7

Introspect Nodes

When registering nodes as described above, if you leave out any of the hardware properties for any of the nodes, introspection will be run on the nodes to obtain these missing properties, as soon as you click the Register Nodes button. In this case, the nodes will be located in the Maintenance tab and will have the status Discovering. After the introspection process has finished (this can take up to 5 minutes for VM / 15 minutes for baremetal), the hardware properties will get populated and the nodes will have the status Discovered. At this point, you can move the nodes to the Free tab, by selecting them using the checkbox on the left side and clicking the Activate Nodes button. Now the nodes are ready and available for deployment.

Create Flavors

To create the necessary flavor, navigate to the Flavors page. One suggested flavor, matching the hardware properties of the created nodes, will be available. To create it, open the dropdown menu under Actions, click Edit before creating, change the name to 'baremetal' and click the Create Flavors button.

Configure Roles

To configure deployment roles, navigate to the Deployment Roles page. Flavor and Image needs to be set to all the deployment roles. For each of the deployment roles, click the edit button and set the Flavor to 'baremetal' and Image to 'overcloud-full'. Save the form.

Service Configuration

To perform the necessary service configuration, navigate to the Service Configuration page and click the Simplified Configuration button. In the Service Configuration form, make sure that the values of the Deployment Type and Public Interface fields are correct. Also make sure you set the SNMP Password and the Cloud name.

Deploy the Overcloud

To deploy the overcloud, navigate to the Overview page. The deployment plan validation will be performed and if the plan is valid, the Verify and Deploy button will be enabled. Click this button to open the deployment confirmation dialog. In case you want to enable network isolation, check the Enable Network Isolation box. Click Deploy.

This will trigger the creation of the overcloud heat stack. The page will reload and you will be able to monitor the current status of the deployment. On the right side you will see the progress bar as well as the last event from the Heat event list. If you want to see the full event list, you can navigate to the Deployment Log page.

Initialize the Overcloud

Once the deployment has successfully completed, you need to perform the initialization of Keystone and Neutron in the overcloud. To do this, click the Initialize button, fill out the form and click Initialize. Once the initialization has completed, the page will reload and you will see deployment details on the Overview page. On the left side the information about roles and node counts will be displayed, along with the system load charts for each deployment role. On the right side, the access information for the overcloud Horizon will be displayed.

Post-Deployment

Access the Overcloud

When the overcloud is deployed, the access information needed to to log into the overcloud Horizon is located on the Overview page.

Redeploy the Overcloud

The overcloud can be redeployed when desired. First, you have to delete the existing overcloud by clicking the Undeploy button on the Overview page. This will trigger the deletion of the Heat stack. After the overcloud has been deleted, the Overview page will again display the deployment checklist along with the Verify and Deploy button. If you wish to deploy the overcloud again, repeat the steps from the Deploy the Overcloud section on this page.