KaTeker Setting up

Topics

Steps setting up KaTeker

Introduction to initial usage of KaTeker

Steps setting up KaTeker

Setting up KaTeker will be made in two steps:

Definition of directories and hardware.
Definition of the data to be saved.

If you are using KaTeker the first time an you got no experience with backup at all you should use KaTeker initially to save the data of a user. This is described in Typical setup saving user data and Introduction to initial usage of KaTeker.

1. Definition of directories and hardware

Definition of the save name

The save name is used to identify the media in order to ensure that it belongs to a backup.

Definition of the directories

KaTeker requires two directories. The save root holds the master and update archives. It is recommended to use for this data a hard disk of its own, not only a partition. This is important when a hard disk crashes. If the data saved and the save are on the same hard disk, both the save and the data cannot be accessed in case of a hard dick crash. On a Linux system it is a good idea to put onto the first hard disk the data and onto the second the save and the swap.

If you want to save data via network onto a server the save root may be mounted via NFS. This is useful if you got no CD recorder on a computer, no second hard disk on it or you want to save the data on a central server.

The second directory is the create root. It is used for temporary i.e. while saving data onto CDs. This directory should be able to hold the CD image of up to 800MB. In many cases /tmp can be used.

Use the Dialog "Setup->Main parameter" to define these directories. This page shows the initial "Setup->Main parameter" dialog.

Selecting the archiver

KaTeker supports two types of archives: tar and dar. While tar is optimized for tape backups dar is optimized for disk backups. The archiver tar is installed on almost every Linux system but not dar. If dar is installed when KaTeker is started the first time, it will prefer dar. The selection can be changed. It can be changed later too. Although KaTeker can handle a mix of dar and tar archives it is better to select the archiver when KaTeker is set up.

Definition of the hardware and the media type

In most cases you'll use the CDR(W) recorder to save data.

A problem can be the right to access the CDR(W) recorder by a nonroot user. This is because of security reasons. It is a little more difficult, because the software used - cdrecord - uses not only the device but a SCSI driver to access the reorder. The solution suggested by cdrecord is to use suid. Another solution is to use sudo to give selected users right to access the recorder. KaTeker supports the both but provides an dialog to setup the sudo rights. The command kateker_cdr_setup may started as root immediately.

KaTeker will search for CD devices initially. Thus setting up the CDR(W) recorder using the "Setup->CDR recorder" dialog is simply to select one of the devices in the list. If the CDR recorder can record CDRW you can select this. CDRs can be used to save data of more than size of of one CD. In this case there will be used multiple CDs. This is useful when saving the master archives. That is the CDRs are multiple. The name of a CDR will be generated automatically when saving data.

On the other side it is not a god idea to save the daily updated data onto multiple CDs because this cannot be done automatically. The best choice is to use CDRWs for this. CDRWs are 'flexible' that is each defined CDRW will be accepted automatically by the daily backup job. CDRW's have to be named and initialized before the can be used. Use the "CDRWs" dialog to define new CDRW media.

When creating a CDR a CD cover label can be generated and printed. This and some information to put onto the CDs are defined in the dialogs "Setup->CDR" and "Setup->CDRW".

The dialog "Setup->CD" can be used to select the CD device. This Device will be used when restoring data. If you defined a CD recorder it will be used to read CDs too, but you can change it.

2. Definition of the data to be saved

The data to be saved should be planned very well. If you save to much data you'll require a lot of backup space on hard disk and save media. But on the other side it is painful when you cannot restore some data lost. One will not save temporary data of course.

Another aspect is to put the data into samples. It is simpler to manage such samples. These samples a the backup sets. KaTeker requires at least one backup set. The members of a backup set are the directories and files to be saved. For each entry in a backup set there will be created a tar or dar archive. That is, a backup set is a sample of tar or dar archives. These archives should not be to big or to small. A size of 10 to 300 Mbyte is recommend. The size has to be less than the size of the save media. KaTeker does not split up tar or dar archives but supports the user to define backup sets with proper archives.

Backup sets are defined by the "Setup->Define backup sets" dialog. A backup set is a list of directories and files to be saved. In each directory included you may exclude some files and directories. It is possible to enter a directory in its parent directory into the backup set immediately. This can be good idea because for each entry in a backup set there will be created a tar or dar archive. This can be necessary too if the size of the associated tar or dar archives are to large to fit on one save media.

Setting up the backup sets can be a lot of work. Starting tkateker initially it will generate backup sets which meets the requirements above. The backup sets are generated when the "<Setup -> Main parameter" are defined the first time.

Finally it is possible to exclude some file names generally using the "Setup->Excludes" dialog.