Backing up tsuru database¶
In the tsuru repository, you will find two useful scripts in the directory
healer.bash. In this page you will
learn the purpose of these scripts and how to use them.
backup.bash uses S3 to store archives, and
downloads archives from S3 buckets. In order to communicate with S3 API, both
scripts use s3cmd.
So, before running those scripts, make sure you have installed s3cmd. You can install it using your preferred package manager. For more details, refer to its download documentation.
After installing s3cmd, you will need to configure it, by running the command:
$ s3cmd --configure
mongodump, creates a tar archive and send
the archive to S3. Here is how you use it:
$ ./misc/mongodb/backup.bash s3://mybucket localhost database
The first parameter is the S3 bucket. The second parameter is the database host. You can provide just the hostname, or the host:port (for example, 127.0.0.1:27018). The third parameter is the name of the database.
Automatically restoring on data loss¶
The other script in the
misc/mongodb directory is
script checks a list of collections and if any of them is gone, download the
last three backup archives and fix all gone collections.
This is how you should use it:
$ ./misc/mongodb/healer.bash s3://mybucket localhost mongodb repositories users
The first three parameters mean the same as in the backup script. From the fourth parameter onwards, you should list the collections. In the example above, we provided two collections: “repositories” and “users”.