Skip to main content

Setting up backups with Bareos

Bareos is an open-source backup solution for client-server environments. Using the droplet storage backend, the storage daemon can write volumes directly to an S3 bucket.

Prerequisites​

  • A running Bareos installation (Director, Storage Daemon, File Daemon)
  • Package for the S3 backend installed, e.g. bareos-storage-droplet
  • S3 access credentials and a Bucket, e.g. bareos-backups

1. Create a droplet profile​

/etc/bareos/bareos-sd.d/device/droplet/centron.profile:

use_https = true
host = s3.internet1.de
access_key = IHR_ACCESS_KEY
secret_key = IHR_SECRET_KEY
pricing_dir = ""
backend = s3
aws_auth_sign_version = 4
aws_region =

Restrict file permissions:

chown bareos:bareos /etc/bareos/bareos-sd.d/device/droplet/centron.profile
chmod 600 /etc/bareos/bareos-sd.d/device/droplet/centron.profile

2. Define the S3 device in the Storage Daemon​

/etc/bareos/bareos-sd.d/device/S3-Device.conf:

Device {
  Name = "S3-Device"
  Media Type = "S3-Object"
  Archive Device = "S3 Object Storage"
  Device Options = "profile=/etc/bareos/bareos-sd.d/device/droplet/centron.profile,bucket=bareos-backups,chunksize=100M"
  Device Type = droplet
  Label Media = yes
  Random Access = yes
  Automatic Mount = yes
  Removable Media = no
  Always Open = no
  Maximum Concurrent Jobs = 1
}

3. Create a storage resource in the Director​

/etc/bareos/bareos-dir.d/storage/S3-Storage.conf:

Storage {
  Name = "S3-Storage"
  Address = "sd.example.com"        # FQDN des Storage Daemons
  Password = "PASSWORT_DES_SD"      # muss zur Director-Ressource im SD passen
  Device = "S3-Device"
  Media Type = "S3-Object"
}

4. Display the pool and job on S3​

Pool {
  Name = "S3-Pool"
  Pool Type = Backup
  Recycle = yes
  AutoPrune = yes
  Volume Retention = 30 days
  Maximum Volume Bytes = 50G
  Label Format = "s3-vol-"
}

Job {
  Name = "backup-server01-s3"
  JobDefs = "DefaultJob"
  Storage = "S3-Storage"
  Pool = "S3-Pool"
}

5. Reload the services and test them​

systemctl restart bareos-sd
systemctl reload bareos-dir

bconsole
* status storage=S3-Storage
* run job=backup-server01-s3 yes
* messages

After the first job, volume objects should appear in the bucket (aws s3 ls s3://bareos-backups/ --endpoint-url https://s3.internet1.de).

Restore​

Restores run as usual via bconsole:

* restore client=server01-fd

Bareos reads the required chunks transparently from the bucket.

Notes​

  • Chunk size: 100 MB is a good standard; larger chunks reduce the number of requests, whilst smaller ones improve parallelism.
  • Do not apply lifecycle rules to the Bareos bucket – retention is controlled by pools/retention in Bareos.
  • For higher throughput, define multiple devices with the same media type and increase Maximum Concurrent Jobs.
Last updated on