ActiveMQ的Kaha Persistence

持久化 http://activemq.apache.org/persistence.html

消息存储 http://activemq.apache.org/amq-message-store.html

Kaha Persistence 基于文件式的本地存储持久化方案，速度最快（现在已经是AMQ默认的持久化方式）可轻松恢复。

Kaha存储数据的方式类似于写日志，它在文件结尾追加记录。消息被存储在单个文件中，文件大小以配置为准，默认是32MB（如果消息本身大于32MB可以修改配置文件）当所有存储在data log中的消息都被成功的消费后，这个日志文件将被标识为待删除或归档，待下一次系统自动清理时将会被清除。

ActiveMQ 5.0 and above:

<broker brokerName="broker" persistent="true" useShutdownHook="false">    <transportConnectors>      <transportConnector uri="tcp://localhost:61616"/>    </transportConnectors>    <persistenceAdapter>      <kahaPersistenceAdapter directory="activemq-data" maxDataFileLength="33554432"/>    </persistenceAdapter>  </broker>

ActiveMQ 4.1 and earlier:

<broker brokerName="broker" persistent="true" useShutdownHook="false">    <transportConnectors>      <transportConnector uri="tcp://localhost:61616"/>    </transportConnectors>    <persistenceAdapter>      <kahaPersistenceAdapter dir="activemq-data" maxDataFileLength="33554432"/>    </persistenceAdapter>  </broker>

如果将属性persistent设置为false，那么消息将采用内存式的存储方式。

每个kahadb实例都可以单独配置，如果没有指定目标队列支持filteredKahaDB，默认值将匹配每一个队列或主题。这将灵活的捕获所有，否则将会因为一个异常而创建失败。

配置方式如下

<broker brokerName="broker" ... >

 <persistenceAdapter>

  <mKahaDB directory="${activemq.base}/data/kahadb">

    <filteredPersistenceAdapters>

      <!-- match all queues -->

      <filteredKahaDB queue=">">

        <persistenceAdapter>

          <kahaDB journalMaxFileLength="32mb"/>

        </persistenceAdapter>

      </filteredKahaDB>

       

      <!-- match all destinations -->

      <filteredKahaDB>

        <persistenceAdapter>

          <kahaDB enableJournalDiskSyncs="false"/>

        </persistenceAdapter>

      </filteredKahaDB>

    </filteredPersistenceAdapters>

  </mKahaDB>

 </persistenceAdapter>

...

</broker>

Automatic per destination persistence adapter

如果filteredKahaDB的perDestination属性设置为true，那么匹配的目标队列将会得到自己对应的KahaDB实例。配置

When the perDestination boolean attribute is set to true on the catch all (no explicit destination set), filteredKahaDB. Each matching destination will get its own kahaDB instance.

<broker brokerName="broker" ... >  <persistenceAdapter>   <mKahaDB directory="${activemq.base}/data/kahadb">     <filteredPersistenceAdapters>       <!-- kahaDB per destinations -->       <filteredKahaDB perDestination="true" >         <persistenceAdapter>           <kahaDB journalMaxFileLength="32mb" />         </persistenceAdapter>       </filteredKahaDB>     </filteredPersistenceAdapters>   </mKahaDB>  </persistenceAdapter> ... </broker>

Note:

Specifying

perDestination="true" and queue=">"

on the same line has not been verified to work and may result in

Reason: java.io.IOException: File '/opt/java/apache-activemq-5.9.0/data/mKahaDB/lock' could not be locked as lock is already held for this jvm.