Skip to content

Latest commit

 

History

History
125 lines (99 loc) · 3.86 KB

English.md

File metadata and controls

125 lines (99 loc) · 3.86 KB
Raw

Vansinetes

Main

README

Requirements before use

  1. Currently works only on amd64 architecture.

  2. In case of Windows operating system, it is recommended to turn off the hypervisor.

bcdedit /set hypervisorlaunchtype off
shutdown -r -t 0
  1. Vagrant and VirtualBox must be installed in the user's environment in advance.

How to Use

Type the following command into the bash shell in the path where the Vagrantfile is located.
For Windows OS, it is recommended to use git bash.

vagrant plugin install vagrant-env && \
vagrant up --no-provision && \
machines=$(vagrant status | tail -8 |  head -n 4 | awk '{ print $1}') && \
for machine in $machines; do
vagrant snapshot save $machine up
done && \
root_pass=vagrant vagrant provision --color && \
vagrant halt && \
for machine in $machines; do
vagrant snapshot save $machine kube
done && \
vagrant up

$ vagrant ssh $(vagrant status | tail -5 | sed -n '1p' | awk '{ print $1}')
password: vagrant

Command meaning

# Install required plugins
$ vagrant plugin install vagrant-env

# Create a snapshot of the initial installation state in case provisioning fails
$ vagrant up --no-provision
$ vagrant snapshot save up

# It can be initialized with the saved snapshot with the following command
$ vagrant snapshot restore up

# If you specify the root_pass=true option, you can specify the root password at the prompt.
$ root_pass=true vagrant provision --color
# Alternatively, you can directly enter the password excluding 'true' and 'yes' in the terminal.
$ root_pass=<password> vagrant provision --color

# For more detailed information, add the debug option.
$ debug=true root_pass=true vagrant provision --color

# Connect to master node
$ vagrant ssh $(vagrant status | tail -5 | sed -n '1p' | awk '{ print $1}')

# If it is the default setting, you can access it with the following command
$ vagrant ssh m1
$ ssh [email protected] # In a Linux environment, a separate host network configuration is required in a virtual box
$ ssh -p 5701 vagrant@localhost

# initial vagrant user password
vagrant

Note

1. Configure the .env, templates/cluster.erb files. (The basic settings are already done)

# .env
MIRROR_CHANGE=yes
# There was a case where an error occurred due to a repo problem in the generic/centos8 image.
# In this case, change this option to yes and refer to the repo files in the files/mirror path and create them.

2. Some tools are commented out to speed up provisioning. Uncomment the necessary parts in ansible/site.yaml.

- name: utils, components install
  hosts: cluster
  roles:
  # - role: utils/k9s
  # - role: utils/stern
  # - role: utils/dashboard
  # - role: utils/builder
  # - role: utils/skopeo
  # - role: utils/compose

3. Several aliases are automatically registered.

alias ans='ansible'
alias anp='ansible-playbook'
alias vi='vim'
alias d='docker'
alias k='kubectl'

4. A private docker registry is created and can be used with initial admin account.(password: admin)
User settings can be set in templates/cluster.erb.

$ curl -u 'admin:admin' https://m1.dev/v2/
{}
$ curl -u 'admin:admin' https://registry.m1.dev/v2/
{}

5. You can use docker and kubectl commands as the vagrant user.

$ d ps
$ k get nodes -o wide

Trouble Shooting

1. If provisioning fails, try again with the vagrant provision command.

2. When deleting and recreating a virtual machine like vagrant destroy --force && vagrant up,
it appears to have been deleted normally, but it is not actually deleted once at a time. Check it yourself, delete it, and try again.