Sharp JX-9500H Instruções de Operação descarregar pdf (Página 14)

Xprint F.A.Q. Document revision 10 (16.09.2008)

Page 14/89

2. Database replication

The replicated database mechanism for the product management is based on the active master/potential master

roles of the hosts in the Xprint domain.

When the active master is down, a potential master takes automatically its place.

Ex: If active master is down, the first Potential Master in the hierarchy will automatically take its place.

1. Enhanced High Availability:

A print job can be considered as made of 2 components:

• the real data to be printed

• the description of how to print the document

These parts can be found at different locations:

• To avoid needless file transfer, the documents to be printed stay on the Client system until printing. So,

the document is at most transferred once to the printing system.

• The job description resides in the "control file" stored in the server directory (i.e. $XPRINT_HOME/SV/…)

on the Server system

A preliminary remark is that the job recovery is only possible if the Xprint directories of the failing host are still

accessible during the crash (for example via a RAID system).

Two major approaches are usually followed to provide high availability:

A) The failing host is replaced completely by a new starting backup system

In the following picture, primary node "HOST A" gathers all critical applications (including the print server)

Backup node "HOST B" is used for non-critical operations

When the primary node stops, the backup node has to automatically:

• shutdown the non critical services

• switch some disks in order to access primary node file systems and recover the critical applications

1 2 ... 9 10 11 12 13 14 15 16 17 18 19 ... 88 89

Sem comentários

Sharp JX-9500H Instruções de Operação Página 14