[Help][Terracotta OS]How Ehcache Client connect to passive/mirror Server on Active Server Failure

[Danish Gondal]

I am new with terracotta open source with ehcache 3.8 .So need some clarity  

Specifying only the active server url in client level code 

How the following client will connect to passive/mirror server when erver active server goes down??

Any change is required in the client code where i am specifying the url ????

or how it will connect to passive/mirror ??

Command to shut down the active server if I want to check make sure passive /mirror got active and start communicate with it  ???? 

Waiting for kind reply 

starting only one server 

F:\torracottaserver\server\bin start-tc-server.bat -f ..\conf\tc-config.xml -n clustered

 

Client Code :

  StatisticsService statisticsService = new DefaultStatisticsService();

    URI uriClusteredOne = create("terracotta://192.168.0.119:9410/clustered");

    try (CacheManager cacheManager = newCacheManagerBuilder()

            .with(cluster(uriClusteredOne).autoCreate().defaultServerResource("default-resource"))

            .withCache("basicCache",

                    newCacheConfigurationBuilder(Long.class, String.class,

                            heap(100).offheap(1, MB).with(clusteredDedicated(5, MB))))

            .using(statisticsService).build(true)) {

      Cache<Long, String> basicCache = cacheManager.getCache("basicCache", Long.class, String.class);

      //LOGGER.info("Putting to cache");

      basicCache.put((long) 1, "Ehsan Gondal");

      //System.out.println("cache retrival"+ basicCache.get(1L));

      //LOGGER.info("Closing cache manager");

    }

Terracotta Server Config file :

<?xml version="1.0" encoding="UTF-8"?>

<tc-config xmlns="http://www.terracotta.org/config"

           xmlns:ohr="http://www.terracotta.org/config/offheap-resource">

  <!--

    This is the default Terracotta server configuration file for the Ehcache kit.

    It defines a single offheap resource of 512MB to be used for caching data.

    It also defines a single server, but you can add another one to benefit from high availability.

  -->

  <plugins>

    <config>

      <ohr:offheap-resources>

        <ohr:resource name="default-resource" unit="MB">512</ohr:resource>

      </ohr:offheap-resources>

    </config>

  </plugins>

  <servers>

    <server host="localhost" name="clustered">

      <!--

        Indicates the location for logs files - %H will resolve to user home directory.

        Note that relative path will be resolved from the location of this configuration file.

       -->

      <logs>%H/terracotta1-logs</logs>

      <!--

        This port is used by clients to communicate to the server.

        Its value is actually the default one and is thus omitted.

      -->

      <tsa-port>9410</tsa-port>

      <!--

        This port is used for server to server communication.

        Its value is actually the default one and is thus omitted.

      -->

      <tsa-group-port>9430</tsa-group-port>

    </server>

<server host="localhost" name="clustered2">

      <!--

        Indicates the location for logs files - %H will resolve to user home directory.

        Note that relative path will be resolved from the location of this configuration file.

       -->

      <logs>%H/terracotta2-logs</logs>

      <!--

        This port is used by clients to communicate to the server.

        Its value is actually the default one and is thus omitted.

      -->

      <tsa-port>9410</tsa-port>

      <!--

        This port is used for server to server communication.

        Its value is actually the default one and is thus omitted.

      -->

      <tsa-group-port>9430</tsa-group-port>

    </server>

    <!--

      Below a sample server definition that will give HA to the cluster if run on a different host.

      Servers know how to communicate between each others because each version of the config file on each host

      will list all servers in it.

    -->

    <!--<server host="otherhost" name="other-server">-->

      <!--<logs>logs</logs>-->

      <!--<tsa-port>9410</tsa-port>-->

    <!--</server>-->

    <!--

      Indicates how much time a server taking over after a failure in an active will wait

      to allow existing clients to reconnect. The time unit is seconds.

    -->

    <client-reconnect-window>120</client-reconnect-window>

  </servers>

  <failover-priority>

    <availability/>

  </failover-priority>

</tc-config>