go-dumper

It's a DB dump tool which will create a backup from source database, then will zip the backup file and upload it to GCS or AWS bucket, finally will restore the data to destination server.

• Creates backup form production database
• Loads backup sql file to staging database

Options

• Usage: dumper [command]

• Available Commands:

  • dump Creates backup form production database
  • load Loads backup sql file to staging database

• Flags:

  • --create-database Create new schema to target DB. Format will be like: dbname_20180101
  • --delete-dump-files Delete Dump files .sql and .tar
  • dump
    • --skip-bucket-store It will ignore bucket storing
    • --skip-restore Only backup will be created
  • load
    • --dump-path It can be either local filepath or GS bucket path

Settings

• To run tool:

  • Run dumper via run.sh:
    • ./run.sh
    • ./run.sh prod [for production]

• To run dumper as a CronJob in kubernets cluster(Production)

  • Update deploy/cronjob.yaml file with your desired time to run
  • Apply the above changes using following command: kubectl apply -f cronjob.yaml
  • Other userful commands
    • kubectl get pods
    • kubectl describe pod
    • kubectl delete jobs --all

Database Support

DB Name	Supported
Mysql	✅
Postgres	❌
Sql Server	❌

Notes:

• important please use mysql client 8.0 or higher to avoid time out issues for large datbase
• used viper to import config.
• used logrus to print logs
  • logrus is completely api-compatible with the stdlib logger, so you can replace your log imports everywhere with log "github.com/sirupsen/logrus"