SQL/MX 2.x Installation and Management Guide (G06.24+, H06.03+)

Managing an SQL/MX Distributed Database
HP NonStop SQL/MX Installation and Management Guide—523723-004
13-8
Using Catalog References in an SQL/MX Distributed
Database
Using Catalog References in an SQL/MX Distributed Database
This subsection describes how to manage catalog references in a Release 2.1 or
Release 2.0 network-distributed database.
System Schema Tables
A record of the catalogs that are registered on a node is maintained in the system
schema on the node. The system schema name,
NONSTOP_SQLMX_nodename.SYSTEM_SCHEMA, enables each node in an
Expand network to have a unique ANSI name for its system schema. The metadata
tables in the system schema are called system schema tables.
Catalog names that begin with NONSTOP_SQLMX_ are reserved for use by NonStop
SQL/MX and cannot be created or registered by users.
Table 13-1 describes the five system schema tables and their roles in catalog
references for Release 2.1 or Release 2.0.
Table 13-1. System Schema Tables
Table Description
ALL_UIDS Represents all object UIDs that are present in metadata tables
on the node and ensures that each UID is unique.
CATSYS Represents catalog visibility on a node. It contains one row for
each catalog that is visible on the node, with an indication of the
schema replication rule (automatic or manual) for the catalog.
Use the CATSYS table to verify whether a catalog is visible on a
node.
CAT_REFERENCES Represents catalog references. The node where a given catalog
is created has the only automatic catalog reference for that
catalog. That node’s table has a row for each catalog reference
for the catalog. Other nodes for which the catalog is registered
have empty manual catalog references for that catalog. Those
nodes’ tables have rows only for the automatic catalog reference
and the local reference. The CAT_REFERENCES table locates
a node with an automatic catalog reference from nodes where
the local reference is empty (manual).
SCHEMA_REPLICAS Represents schema replication. Because Release 2.1 and
Release 2.0 do not support schema replication, schemas for a
catalog are identified only on the automatic catalog reference
node. DDL and utility operations use the SCHEMA_REPLICAS
table to determine what nodes need to be updated as the result
of the operation.
SCHEMATA Represents the schema on a node. It contains one row for every
schema that is visible on the node. If a catalog reference for a
catalog exists on a node and a schema name for a schema in
that catalog is not found in the SCHEMATA table on that node,
that schema does not exist.