Usage

Requirement of Zprem repo

Requirements

1. For Ansible Server (Controller Node) - Minimum 4 CPU, 8GB RAM and 8GB Storage

2. For 0prem Server (Managed Node)

   • For testing purpose - Minimum 16 CPU, 32GB RAM and 50GB Storage

   • Recommended - 48 CPU, 128GB RAM and 150TB Storage

3. The Operating System of Ansible and 0prem Server should be Ubuntu 18.04.6 LTS or Ubuntu 20.04.4 LTS

SSH

Sudo

This documentation presumes that users will employ the sudo utility for elevating privileges from the login user to the root user; a configuration in the file /etc/sudoers such as this:

# Allow members of group sudo to execute any command
sudo   ALL=(ALL:ALL) ALL

Share SSH Keys

Login to the Ansible Server -

a) run the following command in ansible server

sudo su -

after running the above commands your current directory should be /root you can check by using pwd command.

pwd

Before running ansible-playbook the ansible server should share ssh public key to 0prem server.

To create an SSH key:

1. Generate the key:

   ssh-keygen

2. Append the public key (~/.ssh/id_rsa.pub by default) to the file ~/.ssh/authorized_keys of each remote server by using :

   ssh-copy-id [email protected]

   or

   ssh-copy-id user@IPv4

Check the ansible server can login into 0prem server without asking any password by the following command: ssh [email protected] or ssh user@IPv4

0prem

1. Clone the repository on ansible server

   git clone https://github.com/0chain/Zprem.git

2. Change the directory

   cd Zprem/

3. Your current directory should be ~/Zprem and add IPv4 in inventory file NOTE: write same IP in kubespray, master and worker groups

   [kubespray]
   104.34.9.187
   
   [master]
   104.34.9.187
   
   [worker]
   104.34.9.187

4. Edit input.yaml

   server: on-prem    # option [on-prem, aws]
   RemoteUser: root   # write remote server user `Note: use remote user same as used in ssh-copy-id [email protected]`
   cluster: zcn   # write cluster name it will create same name namespace
   domain: 0chain.net    # write your domain name
   sharder: 2    # write sharder count
   miner: 3    # write miner count
   blobber: 6    # write blobber count
   magmacount: 0    # write magma count
   enable_magma: false    # enable magma if want to launch provider
   providercount: 0    # write provider count
   consumercount: 0    # write consumer count
   authorizercount: 0    # write authorizer count
   dns_count: 1           # write 0dns count
   kube_config_path: /root/.0prem/admin.conf    # path where k8s config file present on ansible server
   nodeCount: 1    # Give number of nodes  
   Zminer: false    # do not change it
   Capacity: 1073741824000    # capacity of blobber in bytes
   sharder_tag: staging    # write sharder tag 
   miner_tag: staging      # write miner tag 
   dns_tag: staging      # write 0dns tag 
   blobber_tag: staging    # write blobber tag
   explorer_tag: staging       # write explorer tag 
   block_tag: staging      # write 0block tag 
   box_tag: staging      # write 0box tag 
   proxy_tag: staging      # write 0proxy tag 
   search_tag: staging    # write 0search tag
   blobber_stake_tag: staging    # write blobber-stake tag
   nft_ui_tag: staging    # write 0nft-ui tag
   nft_server_tag: staging    # write 0nft-server tag
   swap_tag: feature-circle-sandbox-e61f3a3d    # write 0swap tag
   readprice: 0.01    # enter read price 
   writeprice: 0.01    # enter write price
   zchain_user: "yes"    #  "yes" for creating account for zchain other-wise "no"
   Admin_Username: admin_Dashboard #  write username [Note - Enter username other then `admin`]
   Admin_Password: password_Dashboard  #  write password [Note - Password must be at least 12 characters]
   slackAPI: https://hooks.slack.com/services/xxxxxxxxx/xxxxxxxxxxx/xxxxxxxxxxxxxxxxxxxxxxxx    # if you want slack messages for you pod OOMKilled issue the enter slack webhook url
   chart_reference: 0chain-dev    # enter helm chart reference
   repos:    
   - name: 0chain-helm            # helm chart repo name
       url: http://0chain-helm-chart.s3-website.us-east-2.amazonaws.com/staging/    # helm chart repo url
   - name: jetstack 
       url: https://charts.jetstack.io
   - name: rancher-stable
       url: https://releases.rancher.com/server-charts/stable
   - name: grafana
       url: https://grafana.github.io/helm-charts
   - name: 0chain-dev
       url: http://0chain-helm-chart.s3-website.us-east-2.amazonaws.com/dev/

5. Install pip3

   sudo apt update && sudo apt install python3-pip -y

6. Install Ansible and other tools

   sudo pip3 install -r requirements.txt

7. Give aws credentials in 0ansible/aws-credentials.yaml file - NOTE: This code itself creates DNS records in Route53

   <cluster>.<domain>

   rancher.<cluster>.<domain>

   grafana.<cluster>.<domain>

Example - DNS Looks like zcn.0chain.net , rancher.zcn.0chain.net , grafana.zcn.0chain.net and 0box.zcn.0chain.net

1. Add Remote User in ansible.cfg file Note: use remote user same as used in ssh-copy-id [email protected]

   [defaults]
   remote_user=root    # edit your remote user
   inventory=inventory    # do not change it
   roles_path=roles    # do not change it
   host_key_checking=False    # do not change it
   ask_pass=False    # do not change it
   ansible_python_interpreter=/usr/bin/python3    # do not change it
   deprecation_warnings=False    # do not change it
   collections_paths=~/.ansible/collections/    # do not change it
   
   [privilege_escalation]
   become=True    # do not change it
   become_method=sudo    # do not change it
   become_user=root    # do not change it
   become_ask_pass=False    # do not change it

Run the Ansible Playbook

Your current directory should be ~/0ansible :

ansible-playbook main.yaml

if you want to check broader logs then run :

ansible-playbook main.yaml -v

Links

This link is created by binding host_address and domain in on_prem.json.

Syntax of URL - https://<cluster_name>.<domain>

1. Explorer - zcn.0chain.net # link of the network you joined [will be shared once whole network is deployed]

2. Rancher - rancher.zcn.0chain.net # link of the rancher you launched

3. Grafana - grafana.zcn.0chain.net # link of the grafana you launched

4. Diagnostics Page - zcn.0chain.net/sharder01/_diagnostics

Update-Resources

After Launching blobbers to a network if you wish to change any value then use update-blobber-resources.yaml file. NOTE: your current path should be ~/0ansible/.

1. Run the following command for updating blobbers value -

ansible-playbook update-blobber-resources.yaml

1. Run the following command for updating miners & sharders value -

ansible-playbook update-miner&sharder-resources.yaml

ResetNetwork

For reset of the network [not the cluster] run reset-0chain.yaml file. NOTE: your current path should be ~/0ansible/.

1. Run the following command for updating value -

ansible-playbook reset-network.yaml

RedeployNetwork

After running the reset playbook run after-reset.yaml playbook. NOTE: your current path should be ~/0ansible/.

1. Run the following command for setting up network again -

ansible-playbook after-reset.yaml

1. A prompt will open which asks for Y/N [here Y is Yes and N is No]. So type Y if you wish to delete configmaps otherwise type N and press ENTER key.

2. About this reset script there are certain points -

Rancher - after redeploy rancher will use password same as used in first deployment [ though there are few ways through which we can create new username & password ]

Grafana - grafana will not pick earlier password after redploying it [ always pick the password given in input file so grafana password can be changed according to set by you ]

Remove-K8SCluster

For removing entire kubernetes cluster change your current directory to /root/kubespray and run the following command -

ansible-playbook -i ~/inventory.yaml  reset.yml