Shulou(Shulou.com)06/02 Report--

Opesntack tranning labs provide an easy way to deploy virtualized OpenStack in desktop environments. Training Labs supports all modern Linux, Windows and macOS platforms. Before running Training Labs, you need to install VirtualBox or KVM / libvirt. Unzip the compressed files into a folder and prepare to deploy your own OpenStack Training Labs cluster with one click.

Download address: docs.openstack.org/training_labs/

Documentation: wiki.openstack.org/wiki/Documentation/training-labs#Building_the_cluster

installation preparation

For installation and test methods, refer to Huawei's HCIP-Cloud_Computing-OpenStack_V1.0_Lab Manual.

Test method, using x86 server, server brand is second-hand HP, hard disk 500G, 4-way CPU, 40G memory, single gigabit network card, address is 172.31.208.31/24. Install unbuntu18.04LTS, use virtualbox 5.2,openstack-labs version is Stein, the following is the recommended experimental environment

Network Description

Experiment environment Install ubuntu operating system on the server, install virtualBox software on ubuntu, deploy openstack experiment environment in virtualbox through openstack automatic installation script, including a controller node and a compute node

ubuntu configuration

After installing ubuntu18 with gnome, we need to update the software source and enable remote desktop login configuration

By default ubuntu does not have a root password, and there is no SSH service. By default, root users are not allowed to log in ssh. need to configure themselves.

sudo passwd #create root password sudo apt-get install opensh-server #install ssh service sudo systemctl start ssh #Open SSHsudo systemctl enable sshsudo systemctl stop ufw #close firewall sudo systemcll disable ufw

In order to speed up software downloads, it is necessary to configure domestic software sources

sudo cp /etc/apt/sources.list /etc/apt/sources.list.bak #Backup software sources sudo gedit /etc/sources.list #Edit

Add at the beginning of the file

deb https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ bionic main restricted universe multiversedeb-src https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ bionic main restricted universe multiversedeb https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ bionic-updates main restricted universe multiversedeb-src https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ bionic-updates main restricted universe multiversedeb https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ bionic-backports main restricted universe multiversedeb-src https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ bionic-backports main restricted universe multiversedeb https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ bionic-security main restricted universe multiversedeb-src https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ bionic-security main restricted universe multiversedeb https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ bionic-proposed main restricted universe multiversedeb-src https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ bionic-proposed main restricted universe multiverse

Update system after saving

sudo apt-get updatesudo apt-get upgrade

Patience waiting for update complete

Next, use xrdp to enable remote login.

sudo apt install xrdp

edit the configuration file

sudo vim /etc/xrdp/startwm.sh

Comment out the bottom two lines of test and exec and add one line

gnome-session

And then ubuntu settings to open sharing

Configure Screen Sharing

After configuration, restart ubuntu. Then you can log in using windows remote desktop

As follows:

Then configure virtualbox

Recommended to use unbuntu gnome software store search download

Confirm installation after download

sudo dpkg-query -l virtualbox

Install opesntack labs

Complete the openstack installation in virtualbox by running the openstack automated installation script.

The controller node and compute node of openstack are both ubuntu operating system (during script installation, the Ubuntu operating system image file will be automatically pulled down, and two virtual machine network cards will be automatically created for ubuntu

vboxnet0:10.0.0.1/24

vboxnet1:203.0.113.1/24

The process is as follows:

First, download openstack, visit docs.openstack.org/training_labs/

After downloading, save to desktop

and then decompress it.

sudo tar -xvf labs-master.tgz

After decompression, the directory structure is as follows:

To avoid insufficient openstack environment, open the config file config.compute1 and modify it as follows

sudo cd labssudo vim labs/config/config.compute1

Install Python

sudo apt install python

Once done, go to the/labs directory and run the st.py script

cd labs./ st.py -b cluster

If there is an error during installation, try running the script again. If there is still a problem, you can view the corresponding log file under/labs/log. After troubleshooting, run the script again.

Wait for the system to automatically install openstack in virtualbox, this process takes about 3 hours. until the following message appears, indicating successful installation

The results that appear represent the default login information for the lab environment.

the problems encountered

During the installation process, you will get a "ERROR vm_ssh:Aborting" card, as follows:

Query the official website documentation, https://wiki.openstack.org/wiki/Documentation/training-labs#Building_the_cluster, below are instructions on using virtualbox to build clusters

The control node and compute node communicate in virtualbox using tcp2230 and 2232 of ssh protocol. I suspect that ssh protocol login does not obtain each other's key.

So I tried the following

ssh -p 2230 osbash@localhost #Password osbash

Then run the openstack installation script again, and it doesn't get stuck here. problem solved

it's amazing

Update:

Later, when deploying, it repeatedly appeared that compute node compute1 could not be installed successfully. This was a bug, and the specific reason was unknown. Later, I found the relevant bug description on foreign forums. The post address is as follows:

https://bugs.launchpad.net/labs/+bug/1803819

I summarized the following, and queried some stackoverflow documents, this is probably python2 subproc module in the shell command call bug, the phenomenon is to use the ssh command will throw an exception, there is a certain probability of occurrence

I'm going to use the following method

That is, manually upload SSH keys and associated configuration files to controller and compute nodes

This is done by:

After the first deployment failure, manually upload the installation script from the host to the controller and compute nodes using ssh when rerunning the installation script.

ssh -q -i /home/yangchao/labs/lib/osbash-ssh-keys/osbash_key -o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no -o ConnectTimeout=10 -o ControlPath=none -p 2230 osbash@127.0.0.1 mkdir -p autostart && ssh -q -i /home/yangchao/labs/lib/osbash-ssh-keys/osbash_key -o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no -o ConnectTimeout=10 -o ControlPath=none -p 2232 osbash@127.0.0.1 mkdir -p autostart

Note that the above command must be run multiple times while the installation script is running, because the controller and compute nodes will restart and configure the virtual network when the deployment script is re-run

After completion, it will be as follows:

At this time, enter http://127.0.0.1:8888/horizon in the browser of the host machine to enter the dashboard of openstack, and at the same time, you can create an instance normally.

Log in to the controller node and check that the nova service is running normally. The compute node has been added.

ssh osbash@127.0.0.1 -p 2230 #Login controller node, password osbash. admin-openrc.sh #Import openstack environment variable to use opnstack command nova service-list #Check the service status of controller node and compute node nova

Remember, only the host can log in to openstack.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.