Anonymization command ​

Considering your anonymization has been configured, you can now anonymize a backup file by running:

console db-tools:anonymize path/to/your/backup/to/anonymized

This command will successively:

1. Backup your local database,
2. Restore the given backup file,
3. Anonymize the data from the given backup file,
4. Backup the newly anonymized database, by overwritting the given backup file,
5. Restore your database to its original state from the backup produced at step 1.

A GDPR-friendly workflow ​

Here is an example of workflow - that follows GDPR recommendations - to retrieve anonymized production data on your local environment.

Prerequisites ​

• You have a second secured environment besides your production (such as a preproduction) and you can securely copy files from one to another,
• You can shut down your service on this second environment,
• Your anonymization is well configured: every sensitive data has been mapped to an anonymizer that will erase/hash/randomize it.

Workflow ​

Let's assume the environment we have besides production is called another_env.

1. Run console db-tools:backup on production environment or choose an existing backup with console db-tools:restore --list,
2. Securely download your backup file from production to another_env environment,
3. Stop services on another_env to ensure no one is using it,
4. Run console db-tools:anonymize path/to/your/production/backup to generate a new backup cleaned from its sensitive data,
5. Restart services on another_env,
6. Download the anonymized backup from another_env to your local machine
7. Restore the backup with console db-tools:restore --filename path/to/your/anonymized/backup

Options ​

You can specify the behaviour of the db-tools:anonymizecommand with some options detailed below.

Anonymizing local database ​

The main purpose of this command is to provide a way to anonymize a backup file. But it could also be used to anonymize local database with --local-database.

console db-tools:anonymize --local-database

Do not restore initial state after anonymization ​

You can choose to not restore initial database with the --no-restore option. With this option, steps 1 and 5 will be skipped during execution.

console db-tools:anonymize --no-restore

Only anonymize specific targets ​

Use this option if you want to anonymize only some specific targets during the process.

console db-tools:anonymize --target target_1 --taget target_2
# or
console db-tools:anonymize -t target_1 -t target_2

Exclude targets from anonymization ​

Use this option if you want to exclude some specific targets from anonymization.

console db-tools:anonymize --exclude target_1 --exclude target_2
# or
console db-tools:anonymize -x target_1 -x target_2

Split update queries ​

By default, the anonymization process use one update query per table. For debug purpose, it could be usefull to run not only one update query per table but one update query per target. To do so, use the --split-per-column option.