Bring up the Control Plane

Bringing up only the control plane

Controller node Configuration

Verify Requirements

Check disk space

Here, we see that there is more than 20GB free in the root file system.

[root@dev01 ~]# df -h /
Filesystem               Size  Used Avail Use% Mounted on
/dev/mapper/cl_dev-root  228G  135G   94G  60% /

Check Available memory

We see that total memory is more than 8G.

[root@dev01 ~]# free -h
              total        used        free      shared  buff/cache   available
Mem:           125G        3.6G        106G         39M         15G        121G
Swap:          4.0G          0B        4.0G

Separate Public and Internal/Admin API Interfaces

Check and configure networking for the Public API

Identify which network interface(s) you plan to use for the public API and ssh. We will refer to this as kolla_external_vip_interface

For this example, we've made the following assumptions, but you will need to customize to your own values.

• kolla_external_vip_interface: eth_public_api

• interface ip address: 192.5.87.10/23

• kolla_external_vip_address: 192.5.87.254

• system hostname: dev01

• Here, we have chosen an interface named eth_public_api to act as the kolla_external_vip_interface

[root@dev01 ~]# ip addr show eth_public_api
2: em1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 44:a8:42:26:ef:aa brd ff:ff:ff:ff:ff:ff
    inet 192.5.87.10/23 brd 192.5.87.255 scope global eth_public_api
       valid_lft forever preferred_lft forever

We see that it has an address configured, and that kolla_external_vip_address is not present. This is correct, as keepalived will manage the VIP address.

Check and configure networking for the Internal/Admin API

Identify which network interface(s) you plan to use for the Admin/Internal API. We will refer to this as network_interface

For this example, we've made the following assumptions, but you will need to customize to your own values.

• network_interface: eth_internal_api

• interface ip address: 10.20.111.10/23

• kolla_internal_vip_address: 10.20.111.254

• system hostname: dev01

• Here, we have chosen an interface named eth_internal_api to act as the kolla_internal_vip_address

[root@dev01 ~]# ip addr show eth_internal_api
2: em1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 44:a8:42:26:ef:aa brd ff:ff:ff:ff:ff:ff
    inet 10.20.111.10/23 brd 10.20.111.255 scope global eth_internal_api
       valid_lft forever preferred_lft forever

We see that it has an address configured, and that kolla_internal_vip_address is not present. This is correct, as keepalived will manage the VIP address.

Check and configure the system hostname

It's important that the system hostname resolves to the interface IP addresses, not to one of the VIPs

[root@dev01 ~]# hostname
dev01
[root@dev01 ~]# getent hosts dev01
10.20.111.254     dev01.chameleoncloud.org dev01

Note that the system hostname should not point to the kolla_internal_vip_address

If it does not resolve to the interface IP, this will be fixed during bootstrap later.

Set up Linux users and permissions

We need somewhere to run the tools, and an account to run them as. We recommend not using the root account to do so. Ensure you have a non-root user account, with paswordless sudo configured. We will refer to this account as deploy-user, and their group as deploy-group. Replace these with your own values.

Ensure permissions and groups

# Create and join deploy-group
[deploy-user@dev01 ~]$ sudo groupadd <deploy-group>
[deploy-user@dev01 ~]$ sudo usermod -a -G <deploy-group> <deploy-user>

# Create and join docker group
[deploy-user@dev01 ~]$ sudo groupadd docker
[deploy-user@dev01 ~]$ sudo usermod -a -G docker <deploy-user>

Log out and back in to refresh group membership, then verify.

[deploy-user@dev01 ~]$ groups
... <deploy-group> docker

Set directory permissions

For this example, we'll be putting files into /opt/, so we need to ensure that it can be read and written by members of deploy-group

[deploy-user@dev01 ~]$ sudo chown root:<deploy-group> /opt
[deploy-user@dev01 ~]$ sudo chmod g+rw /opt

Install Dependencies

Now we need to install some dependencies, then check out and set up chi-in-a-box

# Install Dependencies
sudo apt update && sudo apt install -f -y \
    python3 \
    python3-venv \
    jq \
    git

Site Configuration

Clone and initialize chi-in-a-box. This assumes that we're using the /opt directory that was set up in the last section.

git clone https://github.com/ChameleonCloud/chi-in-a-box /opt/chi-in-a-box
cd /opt/chi-in-a-box
./cc-ansible init --site /opt/site-config

Setting up the ansible hosts file

You'll need to configure a few lines in the file /opt/site-config/inventory/hosts, to tell cc-ansible which services to install where. Here, each of the control,network,monitoring,and storage groups get one entry, dev01, the hostname of your controller node. You don't need to modify the rest of the file.

# /opt/site-config/inventory/hosts

# These initial groups are the only groups required to be modified. The
# additional groups are for more control of the environment.
[control]
dev01

[network]
dev01

[compute]
# No compute node; this is a baremetal-only cluster.

[monitoring]
dev01

[storage]
dev01

[deployment]
localhost ansible_connection=local

# ... Don't edit below here ....

Create a minimal `defaults.yml`

In your site config directory, /opt/site-config, you now have a file called defaults.yml

For a minimal deployment, you only need the following lines uncommented. The values are from the prior section.

# /opt/site-config/defaults.yml

kolla_base_distro: ubuntu
kolla_internal_vip_address: 10.20.111.254
kolla_external_vip_address: 192.5.87.254

network_interface: eth_internal_api
kolla_external_vip_interface: eth_public_api

If you are using a separate deploy host from your controller , then you'll need to put the _interface lines into a separate file instead, /opt/site-config/inventory/host_vars/<hostname>, where hostname is dev01 in this example.

Bootstrap Servers

Run cc-ansible --site /opt/site-config/ bootstrap-servers

This will install docker, configure /etc/hosts, and generally configure the system to run the rest of the installation.

Afterwards, you should see that /etc/hosts contains an entry for your hostname, mapping it to the interface address. Note that this is NOT the haproxy VIP!

$ cat /etc/hosts
...
# BEGIN ANSIBLE GENERATED HOSTS
10.20.111.244 dev01
# END ANSIBLE GENERATED HOSTS

Run Pre-checks

Run cc-ansible --site /opt/site-config/ prechecks

This will warn you about missing configuration, and other common errors.

Commonly seen prechecks errors

Check if ansible user can do passwordless sudo

Make sure that the user executing on the controller node can run sudo without a password. This can be ignored in the case that you're using a separate deploy host, and that the failure is only for the deploy host, not the controller node.

Fail if nscd is running

The nscd service will prevent parts of the deployment from completing successfully. Disable it by running sudo systemctl disable --now nscd.service

Checking host OS distribution / Checking host OS release or version

Ensure that you're running a supported operating system on the controller node. Currently, we support Ubuntu 20.04GeFirst, you'll need to add our docker registry password to the passwords vault.

Downloading Containers

Add our registry password

Run ./cc-ansible --site /opt/site-config edit_passwords

Your default editor will open, and you'll see the following. Ensure that docker_registry_password is configured as follows.

...
docker_registry_password: kaQBdG9PRwlpTzknnhfGmvPf
...

The passwords file will be re-encrypted when you exit the editor.

After configuring this password, pull the containers!

Pull Container images

Run ./cc-ansible --site /opt/site-config pull

Common issues on pull

manifest unknown

If you see an error message like the following, please contact us. It likely means that the container image is missing from our registry, and sometimes happens after CI errors.

404 Client Error for http+docker://localhost/v1.41/images/create?tag=xena&fromImage=docker.chameleoncloud.org%2Fkolla%2Fubuntu-source-keystone: Not Found ("manifest for docker.chameleoncloud.org/kolla/ubuntu-source-keystone:xena not found: manifest unknown: manifest unknown")

Deploy!

You're now ready to run the deploy! This will bring up basic control plane services, listening on the IP addresses you configured above.

Run ./cc-ansible --site /opt/site-config deploy

Access your site

After deploy completes, you'll be able to access the horizon webui at http://<kolla_external_vip_address>

The username is admin, and the password can be found by running cc-ansible --site /opt/site-config view_passwords | grep keystone_admin_password

The next steps will set up easy API access, and create default networks, flavors, and images using the API.

There's not too much useful we can do until we enable some kind of compute service, and tenant network support, so those will come in the next steps.