Deploy On Ansible
Start to Deploy DingoDB
cd dingo-deploy
ansible-playbook playbook.yml
# Offline user
cd dingo-deploy
/usr/local/miniconda3/bin/ansible-playbook playbook.yml
Ansible usage
Command-line usage
cd dingo-deploy
ansible store -m ping # Test all IPs in the [store] group, which can be used to ensure that the Ansible connection is available on each machine before installation
ansible store -m shell -a "ps -ef |grep dingodb_server" # View the dingodb_server process
ansible store -m shell -a "/path/scripts/start-coordinator.sh start/stop/clean/restart/deploy/cleanstart" --become --become-user "myuser" # Start the coordinator node with cleanstart, which includes stop, clean, deploy, and start
ansible store -m shell -a "/path/scripts/start-store.sh start/stop/clean/restart/deploy/cleanstart" --become --become-user "myuser" # Start store node
Management tool DingoControl
# Prerequisite: group_vars/all/_shared.yml and hosts have been configured
# Assign permissions to DingoControl
chmod +x DingoControl
# DiongControl command details
Options:
--help display this help and exit
deploy deploy the application
show show all process for user
stop stop server
clean clean server
start start server
cleanstart stop clean deploy and start server
install install playbook
# install
# Installation options: playbook installs the default configuration, dingodb installs Dingo (executor) and dingo-store, monitor installs all monitoring tools, and the rest are installed in individual steps
# Example: ./DingoControl install playbook
playbook playbook.yml
system install system
jdk install jdk
dingo install dingo
dingo-store install dingo-store
dingodb install dingo and dingo-store
prometheus install prometheus
grafana install grafana
node install node_exporter
process install process_exporter
monitor install prometheus grafana node_exporter process_exporter
# deploy
# The deployment module includes the deployment of the coordinator and the deployment of the store, generating the directory structure before starting
# Example: ./DingoControl deploy all
all stop/clean/deploy store/coordinator
store stop/clean/deploy store
coordinator stop/clean/deploy coordinator
# show
# Example: ./DingoControl show process
process show all process # Display all processes of the deployment user
file show all file not user # Check if there are directories with non-deployment user permissions under the deployment directory
# stop
# Stop processes, you can stop individual processes or stop all of them
# ./DingoControl stop store
all stop all process
store stop store
coordinator stop coordinator
executor stop executor
node-exporter stop node-exporter
process-exporter stop process-exporter
prometheus stop prometheus
grafana stop grafana
# clean
# Clean up files, automatically stop processes. Only 'all' cleans up all files and deletes directories. 'coordinator' and 'store' clean up deployment files and data. The monitoring module cleans up the systemctl service files
# Example: ./DingoControl clean store
all stop all server and clean all file, if want del completely
dingo-store stop/clean store/coordinator and clean dingo-store, deprecated
store stop/clean store
coordinator stop/clean coordinator
node-exporter stop/clean node-exporter
process-exporter stop/clean process-exporter
prometheus stop/clean prometheus
grafana stop/clean grafana
# start
# Start the process, and when starting the store, automatically execute mysql_init.sh to initialize the tables
# Example: ./DingoControl start store
all start all process
store start store
coordinator start coordinator
executor start executor
node-exporter start node-exporter
process-exporter start process-exporter
prometheus start prometheus
grafana start grafana
# cleanstart
# Clear the data and redeploy and start the coordinator, store, and executor
# Example: ./DingoControl cleanstart all
all stop-clean-deploy store/coordinator and restart executor