Bacula 1.29 User's Guide Chapter 1
Index
Index
Next
Features

What is Bacula?

Bacula is a set of computer programs that permit you (or the system administrator) to manage backup, recovery, and verification of computer data across a network of computers of different kinds. In technical terms, it is a network Client/Server based backup program. Bacula is relatively easy to use and efficient, while offering many advanced storage management features that make it easy to find and recover lost or damaged files. Due to its modular design, Bacula is scalable from small single computer systems to systems consisting of hundreds of computers located over a large network.

Who Needs Bacula?

If you are currently using a program such as tar, dump, or bru to backup your computer data, and you would like a network solution, more flexibility, or catalog services, Bacula will most likely provide the additional features you want.

If you are running Amanda and would like a backup program that can write to multiple volumes (i.e. is not limited by your tape drive capacity), Bacula can most likely fill your needs.

If you are currently using a sophisticated commercial package such as Legato Networker. ARCserveIT, Arkeia, or PerfectBackup+, you may be interested in Bacula, which provides many of the same features, and is free software available under the GNU Version 2 software license.

Bacula Components or Services

Bacula is made up of the following five major components or services:
  • Bacula Director service consists of the program that supervises all the backup, restore, verify and archive operations. The system administrator uses the Bacula Director to schedule backups and to recover files. For more details see the Director Services Daemon Design Document. The Director runs as a daemon or a service (i.e. in the background).
  • Bacula Console services is the program that allows the administrator or user to communicate with the Bacula Director (see above). Currently, the Bacula Console is available in two versions. The first and simplest is to run the Console program in a shell window (i.e. TTY interface). Most system administrators will find this completely adequate. The second version is a GNOME GUI interface that for the moment (26 December 2002) is far from complete, but quite functional as it has all the capabilities of the shell Console. For more details see the Bacula Console Design Document.
  • Bacula File services (or Client program) is the software program that is installed on the machine to be backed up. It is specific to the operating system on which it runs and is responsible for providing the file attributes and data when requested by the Director. The File services are also responsible for the file system dependent part of restoring the file attributes and data during a recovery operation. For more details see the File Services Daemon Design Document. This program runs as a daemon on the machine to be backed up, and in some of the documentation, the File daemon is referred to as the Client (for example in Bacula's configuration file). In addition to Unix/Linux File daemons, there is a Windows File daemon (normally distributed as in binary format). The Windows File daemon runs on all currently known Windows versions (95, 98, Me, NT, 2000, XP).
  • Bacula Storage services consist of the software programs that perform the storage and recovery of the file attributes and data to the physical backup media or volumes. In other words, the Storage daemon is responsible for reading and writing your tapes (or other storage media, e.g. files). For more details see the Storage Services Daemon Design Document. The Storage services runs as a daemon on the machine that has the backup device (usually a tape drive).
  • Catalog services are comprised of the software programs responsible for maintaining the file indexes and volume databases for all files backed up. The Catalog services permit the System Administrator or user to quickly locate and restore any desired file. The Catalog services of Bacula set it apart from programs like tar and bru, since the catalog maintains a record of all Volumes used, all Jobs run, and all Files saved. Bacula currently supports three different databases, MySQL, SQLite, and an internal database, one of which must be chosen when building Bacula.

    The internal Bacula database is simple but does not provide the full set of features and queries available with a SQL database. As a consequence, we highly recommend using one of the two SQL databases currently supported (MySQL or SQLite) since they provide many more features, including rapid indexing, arbitrary queries, and security. Although we plan to support other major SQL databases, the current Bacula implementation currently interfaces only to MySQL and SQLite. For more details see the Catalog Services Design Document.

    The RPMs for MySQL ship as part of the Linux RedHat release, or building it from the source is quite easy, see the Installing and Configuring MySQL chapter of this document for the details. For more information on MySQL, please see: www.mysql.com.

    Configuring and building SQLite is even easier. For the details of configuring SQLite, please see the Installing and Configuring SQLite chapter of this document.

To perform a successful save or restore, the following four daemons must be configured and running: the Director daemon, the File daemon, the Storage daemon, and MySQL or SQLite.

Conventions Used in this Document

Bacula is in a state of evolution, and as a consequence, this manual will not always agree with the code. If an item in this manual is preceded by an asterisk (*), it indicates that the particular feature is not implemented. If it is preceded by a plus sign (+), it indicates that the feature may be partially implemented.

If you are reading this manual as supplied in a released version of the software, the above paragraph holds true. If you are reading the online version of the manual, www.bacula.org/manual, please bear in mind that this version describes the current version in development (in the CVS), but generally lags behind the code a bit.

Quick Start

To get Bacula up and running quickly, we recommend that you first scan the Terminology section below, then quickly review the next chapter entitled The Current State of Bacula, then the Quick Start Guide to Bacula, which will give you a quick overview of getting Bacula running. After which, you should proceed to the chapter on Installing Bacula, then How to Configure Bacula, and finally the chapter on Running Bacula.

Terminology

To facilitate communication about this project, we provide here the definitions of the terminology that we use.
Administrator
The person or persons responsible for administrating the Bacula system.
Backup
We use the term Backup to refer to a Bacula Job that saves files.
Catalog
The Catalog is used to store summary information about the Jobs, Clients, and Files that were backed up and on what Volume or Volumes. It permits the administrator or user to determine what jobs were run and their status as well as the important characteristics of each file that was backed up. The Catalog is an online resource, and does not contain the data for the files backed up. Most of the information stored in the catalog is also stored on the backup volumes (i.e. tapes). Of course, the tapes will also have a copy of the file in addition to the File Attributes (see below).

The catalog feature is one part of Bacula that distinguishes it from simple backup and archive programs such as dump and tar.

Client
In Bacula's terminology, the word Client refers to the machine being backed up, and it is synonymous with the File services or File daemon.
Console
The program that interfaces to the Director allowing the user or system administrator to control Bacula.
Daemon
Unix terminology for a program that is always present in the background to carry out a designated task. On Windows systems, as well as some Linux systems, daemons are called Services.
Director
The main Bacula server daemon that schedules and directs all Bacula operations.
Differential
A backup that includes all files changed since the last Full save started. Note, other backup programs may define this differently.
File Attributes
The File Attributes are all the information necessary about a file to identify it and all its properties such as size, creation date, modification date, permissions, etc. Normally, the attributes are handled entirely by Bacula so that the user never needs to be concerned about them. The attributes do not include the file's data.
File Service or File Daemon
The daemon running on the client computer to be backed up. This is also referred to as the File services, and sometimes as the Client services.
FileSet
Defines the files to be backed up. It consists of a list of included files or directories, a list of excluded files, and how the file is to be stored (compression, encryption, signatures). For more details, see the FileSet Resource definition in the Director chapter of this document.
Incremental
A backup that includes all files changed since the last Full, Differential, or Incremental backup started.
Job
A Bacula Job defines the work that Bacula must perform to backup or restore a particular Client. It consists of the Type (backup, restore, verify, etc), the Level (full, incremental,...), the FileSet, and Storage the files are to be backed up (Storage device, Media Pool). For more details, see the Job Resource definition in the Director chapter of this document.
Restore
A restore is the operation of recovering a file (lost or damaged) from a backup medium. It is the inverse of a save, except that in most cases, a restore will normally have a small set of files to restore, while normally a Save backs up all the files on the system. Of course, after a disk crash, a Bacula can be called upon to do a full Restore of all files that were on the system.
Schedule
Defines when the Bacula Job will be scheduled for execution. For more details, see the Schedule Resource definition in the Director chapter of this document.
Service
This is Windows terminology for a daemon -- see above. It is now commonly used in Unix environments as well.
Storage Coordinates
The information returned from the Storage Services that uniquely locates a file on a backup medium. It consists of two parts: one part pertains to each file saved, and the other part pertains to the whole Job. Normally, this information is saved in the Catalog so that the user doesn't need specific knowledge of the Storage Coordinates. The Storage Coordinates include the File Attributes (see above) plus the unique location of the information on the backup Volume.
Session
Normally refers to the internal conversation between the File daemon and the Storage daemon. The File daemon opens a session with the Storage daemon to save a FileSet, or to restore it. A session has a one to one correspondence to a Bacula Job (see above).
Verify
A verify is a job that compares the current file attributes to the attributes that have previously been stored in the Bacula Catalog. This feature can be used for detecting changes to critical system files similar to what Tripwire does. One of the major advantages of using Bacula to do this is that on the machine you want protected such as a server, you can run just the File daemon, and the Director, Storage daemon, and Catalog reside on a different machine. As a consequence, if your server is ever compromised, it is unlikely that your verification database will be tampered with.

Verify can also be used to check that the most recent Job data written to a Volume agrees with what is stored in the Catalog (i.e. it compares the file attributes), *or it can check the Volume contents against the original files on disk.

*Archive
An Archive operation is done after a Save, and it consists of removing the Volumes on which data is saved from active use. These Volumes are marked as Archived, and many no longer be used to save files. All the files contained on an Archived Volume are removed from the Catalog. NOT YET IMPLEMENTED.
*Update
An Update operation causes the files on the remote system to be updated to be the same as the host system. This is equivalent to an rdist capability. NOT YET IMPLEMENTED.
Retention Period
There are various kinds of retention periods that Bacula recognizes. The most important are the File Retention Period, Job Retention Period, and the Volume Retention Period. Each of these retention periods applies to the time that specific records will be kept in the Catalog database. This should not be confused with the time that the data saved to a Volume is valid.

The File Retention Period determines the time that File records are kept in the catalog database. This period is important because the volume of the database File records by far use the most storage space in the database. As a consequence, you must ensure that regular "pruning" of the database file records are done. (See the Console retention command for more details on this subject).

The Job Retention Period is the length of time that Job records will be kept in the database. Note, all the File records are tied to the Job that saved those files. The File records can be purged leaving the Job records. In this case, information will be available about the jobs that ran, but not the details of the files that were backed up. Normally, when a Job record is purged, all its File records will also be purged.

The Volume Retention Period is the minimum of time that a Volume will be kept before it is reused. Bacula will normally never overwrite a Volume that contains the only backup copy of a file. Under ideal conditions, the Catalog would retain entries for all files backed up for all current Volumes. Once a Volume is overwritten, the files that were backed up on that Volume are automatically removed from the Catalog. However, if there is a very large pool of Volumes or a Volume is never overwritten, the Catalog database may become enormous. To keep the Catalog to a manageable size, the backup information should removed from the Catalog after the defined File Retention Period. Bacula provides the mechanisms for the catalog to be automatically pruned according to the retention periods defined.

Scan
A Scan operation causes the contents of a Volume or a series of Volumes to be scanned. These Volumes with the information on which files they contain are restored to the Bacula Catalog. Once the information is restored to the Catalog, the files contained on those Volumes may be easily restored. This function is particularly useful if certain Volumes or Jobs have gone past (that is their retention period and have been pruned or purged from the Catalog. Scanning data from Volumes into the Catalog is done by using the bscan program. See the bscan section of the Bacula Utilities Chapter of this manual for more details.

What Bacula is Not

Bacula is a backup, restore and verification program and is not a complete disaster recovery system unless you plan carefully and follow the instructions included in the Disaster Recovery Chapter of this manual.

With proper planning, as mentioned in the Disaster Recovery chapter Bacula can be used the key component of your disaster recovery system. For example, if you have created an emergency boot disk, a Bacula Rescue disk to save the current partitioning information of your hard disk, and maintain a complete Bacula backup, it is possible to completely recover your system from "bare metal".

If you have used the WriteBootstrap record in your job or some other means to save a valid bootstrap file, you will be able to use it to extract the necessary files (without using the catalog or manually searching for the files to restore).

Interactions Between the Bacula Services

The following block diagram shows the typical interactions between the Bacula Services for a backup job. Each block represents in general a separate process (normally a daemon). In general, the Director oversees the flow of information. It also maintains the Catalog. Interactions between Bacula Services


Index
Index
Next
Features
Bacula 1.29 User's Guide
The Network Backup Solution
Copyright © 2000-2003
Kern Sibbald and John Walker