Introduction

Cloudboostr backup is provided on per component basis. Each component has a separate pipeline for both backup and restore.

Backups bucket

The backup files are being sent to the bucket configured in storage_config and specific config of each deployment and terraform.tfvars. The default config will propagate the name of the bucket from terraform configuration file to all components.

OpsControl main team

To run ELK or Prometheus backup you need to log in to the Concourse using main team credenitals:

1. Log in to the jumpbox and run cb commang to start local container
2. Use following command in that container to get Concourse main team credentials.

   ```bash
   export ENV_NAME="main"
   vault_login
   vault_kv_get_secret /uber_pipeline/${ENV_NAME}/concourse_password
   ```
   

Environment team

To run K8s backup you need to log in to the Concourse using environment administrator credenitals:

1. Log in to the jumpbox and run cb commang to start local container
2. Use following command in that container to get Concourse env team credentials (replace <ENV_NAME> with the name of the environment).

   ```bash
   export ENV_NAME="<ENV_NAME>"
   vault_login
   vault_kv_get_secret /uber_pipeline/${ENV_NAME}/concourse_password
   ```
   

Backup version selection

The backup version can be selected by specifying the backup filename in the backup bucket, namely:

• Kubernetes: ${ENV_NAME}/k8s/k8s_restore_version
• ELK: ${ENV_NAME}/elk/elk_restore_version
• Prometheus: ${ENV_NAME}/prometheus/prometheus_restore_version

The file must contain only the backup file name.

Running backup

The backup procedure is started using specific concourse pipeline described above. The process does not require manual intervention. It can be also configured to run automatically using time_interval property in configuration repository for specific pipeline.