Backup Where

ZMC Backup Where page

Once the backup source for a backup set has been defined on the Backup What page, a backup device must be selected on the Backup Where page. If no devices have been configured (as will be the case with a new installation), the ZMC will take you directly to the Admin Devices page to define one or more backup devices.  If backup devices are defined, Backup Where allows user to define a backup device for the backup set. You will have to provide backup set specific information for each device depending on the type. Sections below provide more information on how to configure the device for the backup set.

The physical object that stores a backup is referred to in Amanda as a medium. Thus, Amanda organizes all backup data in the following heirarchy (note that slots only apply to a tape changer device):

device_medium_diagram.png 

Regardless of the device type, the ZMC also allows you to define an optional write-cache mechanism called a staging area or holding disk, which stores the backup image on the server's hard disk. Because backups can be written in parallel to the holding disk, backups can be completed in much smaller windows than would be possible if writing directly to the device (even if the device is a virtual tape). The staging area can be configured in the Backup Staging page.

Note that while a backup set can include more than one host as a backup source, there can be only one target device per backup set. If you want to send backups from different host systems to different backup media, you must create multiple backup sets to do so. The Duplicate feature of the Admin Backup sets page is useful for this. Also, once a device has been associated to a backup set it cannot be changed (this would invalidate the restore catalog).  To start sending backups from an existing backup set to a new device, duplicate the existing set and bind the resulting backup set to the new device. Note that Backup When settings will have to be manually set to match the old backup set; these settings are not duplicated.

Backup Where Page Overview

This table shows how the devices are used by different backup sets.

BackupWhere-List-3.1.png

Device Name is the name of device as specifiec in Admin Devices page.

Changer Path and Device Path are device file names for tape changer and device. In case of disk backup, Changer Path appears as /dev/null (i.e, changer entry is not valid). In case of Amazon S3 backup, the S3 access key is shown as Device Path.

Media Per Backup defines the number of media volumes that can be used in a backup run. For Amazon S3, tapes and tape changers, the value can be changed in the Backup Where page.

Comments are the notes that were specified when the backup device was configured with the backup set.

 

Advanced Options

Each Device has advanced options that are required to be modified only under special circumstances or when Zmanda Support team requests you to do so. Please see specific device section about the Advanced options available.

Disk backups

Using hard disk space for a backup provides a number of important benefits:

  • The backup window (in other words, the total time to execute backups) can be reduced.
  • Files can be conveniently restored directly from the disk, speeding the restore process.

Disk can be locally attached disk or NAS or SAN device. Disk must accessible from the Amanda server.

 BackupWhere-Disk-3.3.PNG

 Comments

Enter a descriptive comment, such as the physical location of the device or any operational notes.

Root Path

The directory where the backup images will be stored, specified when the device was configured in the Admin Devices page.
Tip: The amount of free space available to hold backup images is an important consideration for fast retrieval of data. To ensure effective restore capabilities, set aside sufficient disk space to hold more than one full Backup Set worth of Data. Just how many full Backup Sets you should keep on disk depends on the data and your sites requirements for quick restores of accidentally deleted data. The more full backup images stored, the longer the retention of accidentally deleted files. Start with enough space to hold three full Backup Set images and adjust this number as experience dictates.
Tip: Because there is no value in creating a backup of a backup on the same media, Zmanda recommends that the drive that holds the vtapes be excluded from the Backup Set that points to the vtapes.

Reserved Percent

Most file systems reserve certain amount of free space. This value is used for free space calculations to provide ZMC warnings when the backup volume is running out of space. 

Advanced Options for Disk should not modified unless Zmanda Support Team recommends you to do so.

Tape Changer

BackupWhere-TapeChanger-3.3.PNG

Comments

Enter a descriptive comment, such as the physical location of the device or any operational notes

Taperscan

The order in which tapes should be searched in the tape changer. Amanda has to load a tape to identify correct tape unless bar code reader is available. 

Auto Label Tapes

The 'Auto Label Tapes' radio button allows users to specify whether they want to use the facility of automatic labeling of tapes. The default value is set to No, meaning ZMC expects you to use the Backup Media page to pre-label the tapes you intended to use for backup. Used tapes cannot be labeled automatically. Amanda will label the tape automatically based on the policy that is specified in the drop-down box. The list of policies are shown below

BackupWhere-AutoLabel-3.1.png

Setting the policy to "Always" is dangerous and is not recommended. If you are using tape changer for multiple backup products including Amanda Enterprise, do not use "Only if Amanda label not found" policy.

Slot range

You can specify the list of slots that can be used for the backup set. Only tapes in these slots will be looked to find tapes for backups. The ranges can be specified as "2-5", "9-13".

Total Tapes Allocated

This value determines the backup retention policy. Amanda will rotate backups among the number of tapes in rotation. You can take tapes out of rotation by archiving it in Backup Media page. Please see Backup When page for more information on this field.

Changer Device

This information is provided for information only and cannot be changed. It can be changed in the Admin Devices page.

Tape Drive Device

The tape drive that should be used by the backup set. Please note that Amanda will use mt command to manage the device. You can specify multiple tape drives for a backup set. This will allow parallel backups to the tape changer.  Users have to match the tape device name to drive slot number. If the match is not done correctly, correct tapes may not be loaded and written to.

Ignore Bar Codes (Advanced Option)

You can disable bar code if the tape do not have bar code labels. This is not recommended because it is difficult to keep track of Amanda media labels and tape.  Users will have track this on their own if the bar codes are not enabled.

On Linux servers, the output of lsscsi and mtx output is provided as a guide to fill out the values as shown below. On Solaris servers, mtx command output is provided (lsscsi command is not available on Solaris).

BackupWhere-TapeChanger-mtxlscsi.PNG

Amazon S3

BackupWhere-S3-3.3.PNG

 

Comments

Enter a descriptive comment, such as  operational notes, amazon account owner.

Location Restriction

The default is US Standard (Closest Amazon US data center). European customers should use European Union(Amazon Ireland data centers) to reduce latency and backup Windows. Other S3 locations can be US-West(Northern California data center), Asia Pacific (Singapore data center) and other S3 locations. The Storage and Data transfer pricing for each S3 location is different.  For details on the storage and data transfer charges associated with different locations, see the Amazon S3 web site at http://aws.amazon.com/s3/.

Backups stored at

The S3 bucket where the backup images will be stored, which was generated when the device was configured in the Admin Devices page.

Secure Communications (Advanced Options)

Enable this to perform secure SSL transfers of data to and from Amazon S3 cloud. Zmanda recommends enabling secure communication.

Speed Per Thread

Users can throttle the network bandwidth for each upload and download thread. The speed limits apply only for the backup set. You can control the backup and restore from the cloud device.

Threads per Upload/Download

Users can control the number of threads used to upload (backup) and download (restore) for the backup set. 

Google Cloud Storage

BackupWhere-Google-3.3.PNG

Comments

Enter a descriptive comment, such as  operational notes, Google account information

Location Restriction

Backups can be stored in US or EU (Europe) in Google Cloud Sotrage

Backups stored at

The  bucket where the backup images will be stored. This value cannot be changed.

Durable Reduced Availability Storage

You can use storage that has less availability. This storage is less expensive. This value cannot be changed after it is set. Changing this value does not migrate data in the cloud.

Secure Communications

Enable or Disable secure backup and recovery to and from the cloud.

Bandwidth throttle

You can use all available bandwidth or throttle the upload (backup) or the download (restore) speed. The throttling is done per thread (see below).

Parallel threads

Each cloud object is uploaded or downloaded in separate thread. You can specify how many threads you want to use. Using too many threads might slow the system down.

 

OpenStack Cloud

 BackupWhereOpenStack.png

Device Name

Unique name for the Device that will be used in further configuration.

Comments

Enter a descriptive comment, such as  operational notes 

Location Restriction

Multiple regions in a OpenStack Cloud

Backups stored at

The  bucket where the backup images will be stored. This value cannot be changed.

Secure Communications

Enable or Disable secure backup and recovery to and from the cloud.

Bandwidth throttle (Speed per Thread)

You can use all available bandwidth or throttle the upload (backup) or the download (restore) speed. The throttling is done per thread (see below).

Parallel threads (threads per Upload/Download)

Each cloud object is uploaded or downloaded in separate thread. You can specify how many threads you want to use. Using too many threads might slow the system down.