WebSphere eXtreme Scale Best Practices for Operation and Management

Chapter 7. Operations and monitoring 167

applications, but on a heavily loaded system, a time-out might occur when no deadlock exists.

Use the lockTimeout attribute to increase the value from the default to prevent false time-out

exceptions from occurring. Set the lockStrategy attribute to NONE to specify that the

BackingMap instance use no lock manager (optional).

For more information about locking strategy, see 6.3, “Locking performance preferred

practices” on page 122.

txTimeout

The txTimeout setting specifies the amount of time in seconds that a transaction is allowed

for completion. If a transaction does not complete in this amount of time, the transaction is

marked for rollback and a TransactionTimeoutException results. If you set the value to 0,

transactions never time out.

Setting requestRetryTimeout on the client

Clients can be configured to retry transactions after a time-out. The time-out value is specified

with the requestRetryTimout property on the client, which is specified in the client properties

file (objectGridClient.properties, by default). If set to a value greater than 0, the request

will be retried on exceptions for which retry is available. Set the value to 0 to fail without retries

on exceptions. The retry logic will run until the requestRetryTimeout (in ms) is reached.

The client will retry a transaction again if it encounters one of the following exceptions:

򐂰 ReplicationVotedToRollbackTransactionException (only on autocommit)

򐂰 TargetNotAvailable

򐂰 org.omg.CORBA.SystemException

򐂰 AvailabilityException (only on autocommit)

򐂰 LockTimeoutException (only on autocommit)

򐂰 UnavailableServiceException (only on autocommit)

The client will

not retry transactions when it encounters one of the following permanent

exceptions:

򐂰 DuplicateKeyException

򐂰 KeyNotFoundException

򐂰 LoaderException

򐂰 TransactionAffinityException

򐂰 LockDeadlockException

򐂰 OptimisticCollisionException

For more information, see this website:

http://publib.boulder.ibm.com/infocenter/wxsinfo/v7r1/topic/com.ibm.websphere.extr

emescale.admin.doc/txsreblmaps.html

7.5 Monitoring WebSphere eXtreme Scale

There are a number of tools and techniques to determine what is happening in a WebSphere

eXtreme Scale environment during run time. In this section, we provide an insight in those

tools and techniques and help you choose which tools and techniques to implement for

monitoring a production environment. We explain the following tools and techniques in this

section:

򐂰 “Operating system monitoring” on page 168

򐂰 “Monitoring WebSphere eXtreme Scale logs” on page 169

򐂰 “WebSphere eXtreme Scale web console” on page 171