org.apache.ignite.spi.failover.jobstealing

Class JobStealingFailoverSpi

java.lang.Object

org.apache.ignite.spi.IgniteSpiAdapter

org.apache.ignite.spi.failover.jobstealing.JobStealingFailoverSpi

All Implemented Interfaces:

FailoverSpi, IgniteSpi

@IgniteSpiMultipleInstancesSupport(value=true)
 @IgniteSpiConsistencyChecked(optional=true)
public class JobStealingFailoverSpi
extends IgniteSpiAdapter
implements FailoverSpi

Job stealing failover SPI needs to always be used in conjunction with JobStealingCollisionSpi SPI. When JobStealingCollisionSpi receives a steal request and rejects jobs so they can be routed to the appropriate node, it is the responsibility of this JobStealingFailoverSpi SPI to make sure that the job is indeed re-routed to the node that has sent the initial request to steal it.

JobStealingFailoverSpi knows where to route a job based on the JobStealingCollisionSpi.THIEF_NODE_ATTR job context attribute (see ComputeJobContext). Prior to rejecting a job, JobStealingCollisionSpi will populate this attribute with the ID of the node that wants to steal this job. Then JobStealingFailoverSpi will read the value of this attribute and route the job to the node specified.

If failure is caused by a node crash, and not by steal request, then this SPI behaves identically to AlwaysFailoverSpi, and tries to find the next balanced node to fail-over a job to.

Configuration

Mandatory

This SPI has no mandatory configuration parameters.

Optional

This SPI has following optional configuration parameters:

• Maximum failover attempts for a single job (see setMaximumFailoverAttempts(int)).

Here is a Java example on how to configure grid with JobStealingFailoverSpi.

 JobStealingFailoverSpi spi = new JobStealingFailoverSpi();

 // Override maximum failover attempts.
 spi.setMaximumFailoverAttempts(5);

 IgniteConfiguration cfg = new IgniteConfiguration();

 // Override default failover SPI.
 cfg.setFailoverSpiSpi(spi);

 // Starts grid.
 G.start(cfg);
 

Here is an example of how to configure JobStealingFailoverSpi from Spring XML configuration file.

 <property name="failoverSpi">
     <bean class="org.apache.ignite.spi.failover.jobstealing.JobStealingFailoverSpi">
         <property name="maximumFailoverAttempts" value="5"/>
     </bean>
 </property>
 

For information about Spring framework visit www.springframework.org

See Also:

FailoverSpi

Field Summary

Fields

Modifier and Type	Field and Description
static int	DFLT_MAX_FAILOVER_ATTEMPTS Maximum number of attempts to execute a failed job on another node (default is 5).

Fields inherited from class org.apache.ignite.spi.IgniteSpiAdapter

ignite, igniteInstanceName

Constructor Summary

Constructors

Constructor and Description
JobStealingFailoverSpi()

Method Summary

Modifier and Type	Method and Description
ClusterNode	failover(FailoverContext ctx, List<ClusterNode> top) This method is called when method ComputeTask.result(org.apache.ignite.compute.ComputeJobResult, List) returns value ComputeJobResultPolicy.FAILOVER policy indicating that the result of job execution must be failed over.
protected List<String>	getConsistentAttributeNames() Returns back a list of attributes that should be consistent for this SPI.
int	getMaximumFailoverAttempts() See setMaximumFailoverAttempts(int).
Map<String,Object>	getNodeAttributes() This method is called before SPI starts (before method IgniteSpi.spiStart(String) is called).
int	getTotalFailedOverJobsCount() Get total number of jobs that were failed over including stolen ones.
int	getTotalStolenJobsCount() Get total number of jobs that were stolen.
JobStealingFailoverSpi	setMaximumFailoverAttempts(int maxFailoverAttempts) Sets maximum number of attempts to execute a failed job on another node.
JobStealingFailoverSpi	setName(String name) Sets SPI name.
void	spiStart(String igniteInstanceName) This method is called to start SPI.
void	spiStop() This method is called to stop SPI.
String	toString()

Methods inherited from class org.apache.ignite.spi.IgniteSpiAdapter

addTimeoutObject, assertParameter, checkConfigurationConsistency0, clientFailureDetectionTimeout, configInfo, createSpiAttributeName, failureDetectionTimeout, failureDetectionTimeoutEnabled, failureDetectionTimeoutEnabled, getExceptionRegistry, getLocalNode, getName, getSpiContext, ignite, initFailureDetectionTimeout, injectables, injectResources, isNodeStopping, onBeforeStart, onClientDisconnected, onClientReconnected, onContextDestroyed, onContextDestroyed0, onContextInitialized, onContextInitialized0, registerMBean, removeTimeoutObject, started, startInfo, startStopwatch, stopInfo, unregisterMBean

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Methods inherited from interface org.apache.ignite.spi.IgniteSpi

getName, onClientDisconnected, onClientReconnected, onContextDestroyed, onContextInitialized

Field Detail

DFLT_MAX_FAILOVER_ATTEMPTS

public static final int DFLT_MAX_FAILOVER_ATTEMPTS

Maximum number of attempts to execute a failed job on another node (default is 5).

See Also:

Constant Field Values

Constructor Detail

JobStealingFailoverSpi

public JobStealingFailoverSpi()

Method Detail

getMaximumFailoverAttempts

public int getMaximumFailoverAttempts()

See setMaximumFailoverAttempts(int).

Returns:

Maximum number of attempts to execute a failed job on another node.

setMaximumFailoverAttempts

@IgniteSpiConfiguration(optional=true)
public JobStealingFailoverSpi setMaximumFailoverAttempts(int maxFailoverAttempts)

Sets maximum number of attempts to execute a failed job on another node. If job gets stolen and thief node exists then it is not considered as failed job. If not specified, DFLT_MAX_FAILOVER_ATTEMPTS value will be used.

Note this value must be identical for all grid nodes in the grid.

Parameters:

maxFailoverAttempts - Maximum number of attempts to execute a failed job on another node.

Returns:

this for chaining.

getTotalFailedOverJobsCount

public int getTotalFailedOverJobsCount()

Get total number of jobs that were failed over including stolen ones.

Returns:

Total number of failed over jobs.

getTotalStolenJobsCount

public int getTotalStolenJobsCount()

Get total number of jobs that were stolen.

Returns:

Total number of stolen jobs.

getNodeAttributes

public Map<String,Object> getNodeAttributes()
                                     throws IgniteSpiException

This method is called before SPI starts (before method IgniteSpi.spiStart(String) is called). It allows SPI implementation to add attributes to a local node. Kernal collects these attributes from all SPI implementations loaded up and then passes it to discovery SPI so that they can be exchanged with other nodes.

Specified by:

getNodeAttributes in interface IgniteSpi

Overrides:

getNodeAttributes in class IgniteSpiAdapter

Returns:

Map of local node attributes this SPI wants to add.

Throws:

IgniteSpiException - Throws in case of any error.

spiStart

public void spiStart(String igniteInstanceName)
              throws IgniteSpiException

This method is called to start SPI. After this method returns successfully kernel assumes that SPI is fully operational.

Specified by:

spiStart in interface IgniteSpi

Parameters:

igniteInstanceName - Name of Ignite instance this SPI is being started for (null for default Ignite instance).

Throws:

IgniteSpiException - Throws in case of any error during SPI start.

spiStop

public void spiStop()
             throws IgniteSpiException

This method is called to stop SPI. After this method returns kernel assumes that this SPI is finished and all resources acquired by it are released.

Note that this method can be called at any point including during recovery of failed start. It should make no assumptions on what state SPI will be in when this method is called.

Specified by:

spiStop in interface IgniteSpi

Throws:

IgniteSpiException - Thrown in case of any error during SPI stop.

failover

public ClusterNode failover(FailoverContext ctx,
                            List<ClusterNode> top)

This method is called when method ComputeTask.result(org.apache.ignite.compute.ComputeJobResult, List) returns value ComputeJobResultPolicy.FAILOVER policy indicating that the result of job execution must be failed over. Implementation of this method should examine failover context and choose one of the grid nodes from supplied topology to retry job execution on it. For best performance it is advised that FailoverContext.getBalancedNode(List) method is used to select node for execution of failed job.

Specified by:

failover in interface FailoverSpi

Parameters:

ctx - Failover context.

top - Collection of all grid nodes within task topology (may include failed node).

Returns:

New node to route this job to or null if new node cannot be picked. If job failover fails (returns null) the whole task will be failed.

getConsistentAttributeNames

protected List<String> getConsistentAttributeNames()

Returns back a list of attributes that should be consistent for this SPI. Consistency means that remote node has to have the same attribute with the same value.

Overrides:

getConsistentAttributeNames in class IgniteSpiAdapter

Returns:

List or attribute names.

setName

public JobStealingFailoverSpi setName(String name)

Sets SPI name.

Overrides:

setName in class IgniteSpiAdapter

Parameters:

name - SPI name.

Returns:

this for chaining.

toString

public String toString()

Overrides:

toString in class Object