In this page you will find documentation on how to mount a bucket as a normal file system through s3ql in user space.

S3QL has features like compression, encryption, data de-duplication, immutable trees and snapshotting which make it especially suitable for online backup and archival.

Using s3ql a  bucket can be mounted using the S3 protocol on linux systems.

Installation

Documentation on how to install s3ql can be found at: https://github.com/s3ql/s3ql/ and https://www.rath.org/s3ql-docs/installation.html. There is also a user guide: http://www.rath.org/s3ql-docs/index.html.

Configuration

Create a file ~/.s3ql/authinfo2 with the following content:

[s3]
backend-login: <access key>
backend-password: <secret keys>
storage-url: s3c://objectstore.surf.nl:443/<bucket>
fs-passphrase: <password>

By default, the data will be encrypted. The fs-passphrase is the password used to encrypt the randomly generated master key. This makes sure that s3ql encrypts data before it uploads the data to SWIFT.

The <container name> and <bucket> are the names of the container that is to be mounted. They are the same.

Don’t forget:

chmod 600 ~/.s3ql/authinfo2

Mounting the file system

If you haven’t mounted the container before, you first have to create a file system. You can do this, for example, in the following way:

mkfs.s3ql s3c://objectstore.surf.nl/<bucket>

If you don’t want to encrypt the file system you can use the commandline flag –plain, as in:

mkfs.s3ql --plain s3c://objectstore.surf.nl/<bucket>

Create a directory that is to be mounted

mkdir /path/to/mount

Mount the object storage

mount.s3ql s3c://objectstore.surf.nl:443/<bucket> /path/to/mount

Unmounting the file system

Unmounting the file system is done by:

umount.s3ql /path/to/mount
  • No labels