Using WebLogic Server Clusters

 

1. Overview of WebLogic Server Clustering

What is a WebLogic Server Cluster?

What Services are Clusterable?

HTTP session state clustering

Object clustering

JDBC connections in a cluster

Non-clustered services and APIs

High-Level Cluster Features

Cluster-Wide JNDI Tree

Load Balancing

Round-robin (default)

Weight-based round-robin

Random

Parameter-based routing

Fail-over

Idempotent objects

HTTP Session State Replication

2. How Clusters Work

Overview

Server Communication in a Cluster

One-to-many communication using IP multicast

Implications for cluster planning and configuration

Peer-to-peer communication using IP sockets

Pure-Java versus native sockets implementations

Configuring reader threads for Java socket implementations

Client communication via sockets

Cluster-wide JNDI naming service

Creating the cluster-wide JNDI tree

Handling JNDI naming conflicts

Updating the JNDI tree

Client interaction with the cluster-wide JNDI tree

Server Failure Detection

Failure detection using IP sockets

The WebLogic Server "heartbeat"

How Servlet Clustering Works

Requirements for servlet clustering

Session Requirements

Accessing clustered servlets

Failover procedure

Using URL re-writing to track session replicas

How Object Clustering Works

Replica-aware RMI stubs

Replica-aware EJB stubs

EJB Home stubs

Stateless EJBs

Stateful EJBs

Entity EJBs

RMI Object stubs

Optimizing for co-located services

Transactional co-location

3. Planning WebLogic Server Clusters

Overview

Capacity planning reminder

Definition of terms

Web application "tiers"

De-Militarized Zone (DMZ)

Load balancer

Proxy plug-in

Planning by Dividing Application Tiers

Limitations of "combined tier" cluster

Load balancing limitations

HTTP session states

Clustered objects

Failover limitations

Security limitations

Addressing security limitations

Recommended 2-Tier Cluster Architecture

Physical hardware and software layers

Web layer

Servlet/Object layer

Benefits of basic configuration

Load balancing and failover for servlets

Improved security options

Limitations of basic configuration

FIrewall limitations

Recommended Multi-tier Cluster Architecture

Physical hardware and software layers

Web layer

Presentation layer

Object layer

Benefits of multi-tier architecture

Load balancing for clustered object calls

Improved Server Load Balancing

Higher availability

Limitations of multi-tier architecture

Firewall restrictions

Configuration notes for multi-tier architecture

Improving Security in Recommended Cluster Architectures

Basic firewall for untrusted connections

DMZ with basic firewall configurations

Combining firewall with load balancer

Expanding the firewall for internal clients

Additional security for shared databases

DMZ with 2-firewall configuration

4. Administering WebLogic Clusters

Overview

Requirements

Network requirements

Licensing

weblogic.properties files

Server Configuration

Configuring a WebLogic Server Cluster

Setting up the Network

Setting up the weblogic.properties files

weblogic.properties file contents

Setting CLASSPATH

Setting up HTTP servlets

Using In-Memory State Replication with a Proxy Server

Setting up the Proxy Server

Setting up the WebLogic Cluster

Using JDBC Session Persistence

Setting Up Enterprise JavaBeans

Starting the WebLogic Server and Joining a Cluster

Starting with Defaults

Details on Starting WebLogic in a Clustered Environment

Command Line Properties

Advanced Properties

Per-Server Properties

Stopping a Server in a Cluster

Special Considerations for Clients of a Cluster

A. Cluster Quick Start

Checklist A-1

Obtain a Cluster-Enabled License for WebLogic Server A-1

Obtain an IP Address for each Server to be Started in the Cluster A-1

Obtain a Multicast Address for Intra-Cluster Communications A-2

Determine The Server Connection Port A-2

Install WebLogic Server into a (Shared) File System in the Cluster A-3

Customize WebLogic Server for a Cluster A-3

Create and Customize a Startup Script for each Server in the Cluster A-4

Start the Individual Servers in the Cluster A-5

B. Troubleshooting WebLogic Server Clusters

Applying Service Packs B-1

Collecting Diagnostic Information B-1

Providing diagnostics to BEA Technical Support B-2

Addressing Common Problems B-3

Tuning client connection timeouts with TIME_WAIT B-3

Server fails to join a cluster B-4

C. The WebLogic Cluster API

How to use the API C-1