Frequently Asked Questions (FAQ) and additional notes

This section is developed as FAQ, and is made up of a list of common questions that could arise for a LIBSAFE administrator in the normal use of the tool.

The conceptual order of the questions tries to follow the same order as the previous sections of this manual.

Is there any limit as for how many user groups may be assigned to a user?

There is no theoretical limitation. Nevertheless, as a reference limit, a big deployment project would rarely need to define user groups for a quantity of more than two digits, and to assign more than a dozen groups to a single user.

How should users be named at creation time?

The most obvious recommendation is to follow the same user name used in the AD/LDAP, if this is being used; if not, a typical schema of name and surname initials, plus a digit number to solve coincidences, is a good idea.

In any case, the user name should be linked always to the same physical person, even if this person stops to use the system. The user name is a historical piece of data kept by the platform, and relevant for the object preservation. This is the reason why users are not to be deleted; and it is neither recommended to de-assign them from the user groups. To set the user to inactive status should be enough.

How to monitor a disk or storage group for exceeding a certain usage limit?

There are alarms defined for this for both entities.

I have a license for 25Tb, but actually only have 10Tb of storage. Is it a problem?

No, it is no problem. LIBSAFE allows configuring disks at any moment, and the platform needs not be run from the first moment with its complete and final storage configuration.

A storage disk may be in trouble, and there is a need to change it / I want to change a storage disk for a bigger one. What should I do?

Once a disk has been defined in the system, the only way to disassemble it from the system is to use the Datamover tools, which will move any object information kept in one disk to a new one, guaranteeing all aspects of system integrity. Remember that the Datamover tools have a section to register the new disk; so do not predefine the substitution disk in the disk section.

Is it possible to use RAID configurations for an improved access speed?

LIBSAFE considers only ‘normal disks’. If the hardware structure of a disk, which is accessed through a local unit or UNC path, has a RAID structure –with ‘data striping’ functionality- LIBSAFE is not even going to know. So, is it a good idea to user these configurations for higher speed? The preservation theory, according to the OAIS standards, sets as a security principle that it has to be possible to access the contents of a repository directly, without the application that has supported its creation and management. Following this point, if the simple access to all RAID structure is guaranteed, it would be a good idea to use this RAID. But if in case of catastrophic event access would be possible only to individual disks, in which objects are not fully individually contained, this kind of configuration should be rejected.

Is it necessary to manage database backup?

LIBSAFE does not manage the database, but just uses it. From this point of view, it is recommended to follow normal organization procedures. Backing up the LIBSAFE database is recommended. But this backup copy should be synchronized with the real storage, since if in case of a fault a database is restored without matching exactly the disk contents, the audits would throw multiple conflict alarms.