Network Config  «Prev  Next»

Lesson 4 When to use Oracle Names
Objective When is it appropriate to use Oracle Names.

Name Application Suitability

Question: Is the "Names Server" which was used for "Oracle Names" in Oracle 8i still used for the successor technology "Directory Naming"?
The "Names Server" used for "Oracle Names" in Oracle 8i is not directly used for the successor technology "Directory Naming." While the concept of a directory service remains, the implementation is entirely different.
Here's why:
Oracle Names:
  • Used a proprietary naming scheme and server architecture.
  • Relied on the "tnsnames.ora" file for configuration.
  • Offered limited functionality and lacked scalability.
  • Was deprecated in Oracle 9i in favor of Net Service Names.

Directory Naming:
  • Leverages industry-standard Lightweight Directory Access Protocol (LDAP).
  • Integrates seamlessly with other directory services like Active Directory.
  • Provides enhanced features like dynamic service registration and discovery.
  • Supports multi-platform environments and boasts improved scalability.
While some fundamental concepts might overlap, the underlying technology and implementation are vastly different. Oracle Names utilized a dedicated server process and file-based configuration, while Directory Naming relies on LDAP and integrates with existing directory services. Here are some key differences:
  1. Protocol: Oracle Names used its own protocol, while Directory Naming uses LDAP.
  2. Integration: Oracle Names operated independently, while Directory Naming integrates with other directory services.
  3. Scalability: Oracle Names had limited scalability, while Directory Naming is highly scalable.
  4. Functionality: Oracle Names offered basic features, while Directory Naming provides advanced features.

Therefore, although both involve directory services, they are distinct technologies with different functionalities and implementations.

Explain when it is and is not appropriate to use Oracle Names.
Oracle Names has some bugs and anomalies that make it unsuitable for some types of Net8 environments. Let us explore the issues.
The most obvious problem is that the product introduces a single point of failure. If the Names server fails for any reason, the entire Oracle network is crippled and all connectivity is lost. Without Oracle Names, each PC independently resolves services names with its tnsnames.ora file. Oracle Names has addressed this issue by allowing the definition of multiple Oracle Names servers that can dynamically share information with one another.
Another problem is the nascent technology of Oracle Names. As Oracle Names continues to evolve, it will undoubtedly become more robust, but within Oracle8 the Oracle Names product still exhibits some unreliability. Of course, this will become less of an issue as Oracle Names matures. Oracle Names is ideally suited to environments that:

  1. Have very reliable network connections
  2. Are very dynamic with new databases being added and frequently changed
  3. Are willing to risk a single point of failure in the network
Oracle Names is not suited to Net8 sites that have a very static database environment where databases rarely change.
Let us look now at how to configure a names.ora file that will store all of the information required for an Oracle Names server.