You are looking at documentation for an older release. Not what you want? See the current release documentation.
This section will show you how to use and configure Infinispan in the clustered environment. Also, you will know how to use a template-based configuration offered by eXo JCR for Infinispan instances.
For indexer, lock manager and data container
Each mentioned components uses instances of Infinispan product for caching in clustered environment. So every element has its own transport and has to be configured in a proper way.
eXo JCR offers a template-based configuration for Infinispan instances. You can have one template for Lock Manager, one for Indexer and one for
data container and use them in all the workspaces, defining the map of substitution parameters in a main configuration file. Just simply define
${infinispan-<parameter name>} inside xml-template and list correct value in JCR configuration file just below
"infinispan-configuration", as shown:
<jgroups transport="org.infinispan.remoting.transport.jgroups.JGroupsTransport">
<stack-file name="stack" path="${jgroups-configuration}"/>
</jgroups>
and JCR configuration file:
...
<property name="infinispan-configuration" value="conf/standalone/cluster/test-infinispan-lock.xml" />
<property name="jgroups-configuration" value="udp-mux.xml" />
...
JGroups is used by Infinispan for network communications and transport in a clustered environment.
If the property jgroups-configuration is defined in component configuration,
it will be injected into the configuration of the Infinispan instance on startup.
<property name="jgroups-configuration" value="your/path/to/modified-udp.xml" />
To prevent having to deal with several different cluster name and to reduce the network overheads, it is now highly recommended to use the JGroups
shared transport, to do so simply set the property singleton_name of your JGroups configuration to a unique name as below:
<config xmlns="urn:org:jgroups"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="urn:org:jgroups http://www.jgroups.org/schema/JGroups-3.2.xsd">
<UDP
singleton_name="JCR-cluster"
...
Shipped Infinispan configuration templates
eXo JCR implementation is shipped with ready-to-use Infinispan configuration templates for JCR's components.
They are located in the application package inside the folder /conf/portal/cluster.
Data container template
Data container template is cache-config.xml:
<infinispan xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xmlns="urn:infinispan:config:8.2"
xsi:schemaLocation="urn:infinispan:config:8.2 http://www.infinispan.org/schemas/infinispan-config-8.2.xsd">
<threads>
<thread-factory name="infinispan-factory" group-name="infinispan" thread-name-pattern="%G %i" priority="5"/>
<!-- listener-executor -->
<blocking-bounded-queue-thread-pool name="infinispan-listener" thread-factory="infinispan-factory"
core-threads="1" max-threads="5" queue-length="0" keepalive-time="0"/>
<!-- expiration-executor -->
<scheduled-thread-pool name="infinispan-expiration" thread-factory="infinispan-factory" />
</threads>
<jgroups transport="org.infinispan.remoting.transport.jgroups.JGroupsTransport">
<stack-file name="stack" path="${exo.jcr.cluster.jgroups.config}"/>
</jgroups>
<cache-container name="jcr" default-cache="default" expiration-executor="infinispan-expiration" listener-executor="infinispan-listener"
statistics="true">
<jmx duplicate-domains="true" domain="jcr.ispn.cache" mbean-server-lookup="org.infinispan.jmx.PlatformMBeanServerLookup"/>
<transport cluster="${exo.cluster.partition.name}-jcr" stack="stack" lock-timeout="240000"/>
<replicated-cache-configuration mode="SYNC" name="default" statistics="true" remote-timeout="${exo.jcr.cluster.workspace.sync.repltimeout:240000}">
<locking isolation="READ_COMMITTED" concurrency-level="500" striping="false" write-skew="false"
acquire-timeout="${exo.jcr.workspace.lockacquisitiontimeout:180000}"/>
<transaction locking="PESSIMISTIC" mode="NON_XA"
transaction-manager-lookup="org.exoplatform.services.transaction.infinispan.JBossStandaloneJTAManagerLookup" />
<eviction size="${exo.jcr.workspace.eviction.maxEntries:1000000}" strategy="LIRS" thread-policy="DEFAULT"/>
<state-transfer enabled="${exo.jcr.cluster.workspace.statetransfer.fetchinmemorystate:false}"
timeout="${exo.jcr.cluster.workspace.statetransfer.timeout:240000}" />
<expiration interval="${exo.jcr.workspace.expiration.wakeupinterval:60000}"/>
</replicated-cache-configuration>
</cache-container>
</infinispan>
| Variable | Description |
|---|---|
exo.jcr.cluster.jgroups.config | This is the path to JGroups configuration that should not be anymore jgroups' stack definitions but a normal jgroups configuration format with the shared transport configured by simply setting the jgroups property singleton_name to a unique name (it must remain unique from one portal container to another). This file is also pre-bundled with templates and is recommended for use. |
{exo.cluster.partition.name}-jcr | This defines the name of the cluster. Needs to be the same for all nodes in a cluster in order to find each other. |
Lock manager template
It's template name is lock-config.xml:
<infinispan xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xmlns="urn:infinispan:config:8.2"
xsi:schemaLocation="urn:infinispan:config:8.2 http://www.infinispan.org/schemas/infinispan-config-8.2.xsd">
<threads>
<thread-factory name="infinispan-factory" group-name="infinispan" thread-name-pattern="%G %i" priority="5"/>
<!-- listener-executor -->
<blocking-bounded-queue-thread-pool name="infinispan-listener" thread-factory="infinispan-factory" core-threads="1"
max-threads="5" queue-length="0" keepalive-time="0"/>
</threads>
<jgroups transport="org.infinispan.remoting.transport.jgroups.JGroupsTransport">
<stack-file name="stack" path="${exo.jcr.cluster.jgroups.config}"/>
</jgroups>
<cache-container name="lock-manager" default-cache="default" listener-executor="infinispan-listener"
statistics="true">
<jmx duplicate-domains="true" domain="jcr.ispn.cache" mbean-server-lookup="org.infinispan.jmx.PlatformMBeanServerLookup"/>
<transport cluster="${exo.cluster.partition.name}-jcr-lock" stack="stack" lock-timeout="240000"/>
<replicated-cache-configuration mode="SYNC" name="default" statistics="true" remote-timeout="${exo.jcr.cluster.lock.sync.repltimeout:240000}">
<locking isolation="READ_COMMITTED" concurrency-level="500" striping="false" write-skew="false"
acquire-timeout="${exo.jcr.lock.lockacquisitiontimeout:180000}"/>
<transaction transaction-manager-lookup="org.exoplatform.services.transaction.infinispan.JBossStandaloneJTAManagerLookup" mode="NON_XA"/>
<state-transfer enabled="${exo.jcr.cluster.lock.statetransfer.fetchinmemorystate:false}"
timeout="${exo.jcr.cluster.lock.statetransfer.timeout:240000}"/>
<eviction strategy="NONE" />
<expiration lifespan="-1" />
<persistence passivation="false">
<string-keyed-jdbc-store xmlns="urn:infinispan:config:store:jdbc:8.0" shared="true" fetch-state="true"
read-only="false" purge="false" preload="true">
<string-keyed-table drop-on-exit="${infinispan-cl-cache.jdbc.table.drop}" create-on-start="${infinispan-cl-cache.jdbc.table.create}"
prefix="${infinispan-cl-cache.jdbc.table.name}">
<id-column name="${infinispan-cl-cache.jdbc.id.column}" type="${infinispan-cl-cache.jdbc.id.type}" />
<data-column name="${infinispan-cl-cache.jdbc.data.column}" type="${infinispan-cl-cache.jdbc.data.type}" />
<timestamp-column name="${infinispan-cl-cache.jdbc.timestamp.column}" type="${infinispan-cl-cache.jdbc.timestamp.type}" />
</string-keyed-table>
</string-keyed-jdbc-store>
</persistence>
</replicated-cache-configuration>
</cache-container>
</infinispan>
To prevent any consistency issue regarding the lock data please ensure that your cache store is org.exoplatform.services.jcr.infinispan.JdbcStringBasedCacheStore and that your database engine is transactional.
| Variable | Description |
|---|---|
exo.jcr.cluster.jgroups.config | This is the path to JGroups configuration that should not be anymore jgroups' stack definitions but a normal jgroups configuration format with the shared transport configured by simply setting the jgroups property singleton_name to a unique name (it must remain unique from one portal container to another). This file is also pre-bundled with templates and is recommended for use. |
{exo.cluster.partition.name}-jcr-lock | This defines the name of the cluster. Needs to be the same for all nodes in a cluster in order to find each other. |
infinispan-cl-cache.jdbc.table.name | The table name. |
infinispan-cl-cache.jdbc.id.column | The name of the column id. |
infinispan-cl-cache.jdbc.data.column | The name of the column data. |
infinispan-cl-cache.jdbc.timestamp.column | The name of the column timestamp. |
infinispan-cl-cache.jdbc.id.type | The type of the column id. |
infinispan-cl-cache.jdbc.data.type | The type of the column data. |
infinispan-cl-cache.jdbc.timestamp.type | The type of the column timestamp. |
infinispan-cl-cache.jdbc.table.drop | Can be set to true or false. Indicates whether to drop the table at stop phase. |
infinispan-cl-cache.jdbc.table.create | Can be set to true or false. Indicates whether to create table at start phase. If true, the table is created if it does not already exist. |
Query handler (indexer) template
Have a look at indexer-config.xml:
<infinispan xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xmlns="urn:infinispan:config:8.2"
xsi:schemaLocation="urn:infinispan:config:8.2 http://www.infinispan.org/schemas/infinispan-config-8.2.xsd">
<threads>
<thread-factory name="infinispan-factory" group-name="infinispan" thread-name-pattern="%G %i" priority="5"/>
<!-- listener-executor -->
<blocking-bounded-queue-thread-pool name="infinispan-listener" thread-factory="infinispan-factory"
core-threads="1" max-threads="5" queue-length="0" keepalive-time="0"/>
<!-- expiration-executor -->
<scheduled-thread-pool name="infinispan-expiration" thread-factory="infinispan-factory" />
</threads>
<jgroups transport="org.infinispan.remoting.transport.jgroups.JGroupsTransport">
<stack-file name="stack" path="${exo.jcr.cluster.jgroups.config}"/>
</jgroups>
<cache-container name="indexer" default-cache="default" expiration-executor="infinispan-expiration" listener-executor="infinispan-listener"
statistics="true">
<jmx duplicate-domains="true" domain="jcr.ispn.cache" mbean-server-lookup="org.infinispan.jmx.PlatformMBeanServerLookup"/>
<transport cluster="${exo.cluster.partition.name}-jcr-indexer" stack="stack" lock-timeout="240000"/>
<replicated-cache-configuration mode="SYNC" name="default" statistics="true" >
<eviction strategy="NONE"/>
<transaction mode="NON_XA" transaction-manager-lookup="org.exoplatform.services.transaction.infinispan.JBossStandaloneJTAManagerLookup"/>
<locking isolation="READ_COMMITTED" concurrency-level="500" striping="false" write-skew="false"
acquire-timeout="${exo.jcr.cluster.indexer.sync.repltimeout:180000}"/>
<state-transfer enabled="${exo.jcr.cluster.indexer.statetransfer.fetchinmemorystate:false}"
timeout="${exo.jcr.cluster.indexer.statetransfer.timeout:240000}" />
<persistence passivation="false">
<store class="${infinispan-cachestore-classname}" shared="false" fetch-state="true" read-only="false" purge="false" preload="false"></store>
</persistence>
</replicated-cache-configuration>
</cache-container>
</infinispan>
| Variable | Description |
|---|---|
exo.jcr.cluster.jgroups.config | This is the path to JGroups configuration that should not be anymore jgroups' stack definitions but a normal jgroups configuration format with the shared transport configured by simply setting the jgroups property singleton_name to a unique name (it must remain unique from one portal container to another). This file is also pre-bundled with templates and is recommended for use. |
{exo.cluster.partition.name}-jcr-indexer | This defines the name of the cluster. Needs to be the same for all nodes in a cluster in order to find each other. |