Server User Manual

For more information on conguring the load balancer plug-in, see “Conguring the HTTP
Load Balancer” in Sun GlassFish Enterprise Server 2.1 High Availability Administration Guide
.
HADB timeouts
The sql_client time out value may aect performance.
Operating System Conguration
If the number of semaphores is too low, HADB can fail and display this error message:
No space left on device
This can occur either while starting the database, or during run time.
To correct this error, congure semaphore settings. Additionally, you may need to congure
shared memory settings. Also, adding nodes can aect the required settings for shared memory
and semaphores. For more information, see
“Conguring Shared Memory and Semaphores” in
Sun GlassFish Enterprise Server 2.1 High Availability Administration Guide
.
Tuning the Enterprise Server for High-Availability
This section discusses how you can congure the high availability features of Enterprise Server.
This section discusses the following topics:
■
“Tuning Session Persistence Frequency” on page 117
■
“Session Persistence Scope” on page 118
■
“Session Size” on page 118
■
“Checkpointing Stateful Session Beans” on page 119
■
“Conguring the JDBC Connection Pool” on page 119
■
Descriptor conguration in the web application
To ensure highly available web applications with persistent session data, the high availability
database (HADB) provides a backend store to save HTTP session data. However, there is a
overhead involved in saving and reading the data back from HADB. Understanding the
dierent schemes of session persistence and their impact on performance and availability will
help you make decisions in conguring Enterprise Server for high availability.
In general, maintain twice as many HADB nodes as there are application server instances. Every
application server instance requires two HADB nodes.
Tuning the Enterprise Server for High-Availability
Sun GlassFish Enterprise Server 2.1 Performance Tuning Guide • January 2009116