Page Comparison

...

Minimum High Availability Deployment for WSO2 IS Analytics

...

1. Do the following database-related configurations.

   1. Follow the steps below to configure the <IS Analytics_HOME>/repository/conf/datasources/master-datasources.xml file as required. 

      Note

      Note that you can point all these datasources to a single database as it is not technically neccessary to separate the data into different databases. However, if required, you can have separate databases as well. The steps given below demonstrate the flow assuming you have created separate databases for each. If you are using a single database instead, simply point the datasources indicated below to a single database.

      1. Enable all the nodes to access the users database by configuring a datasource to be used by user manager as shown below. 

         Code Block
         language xml
         <datasource> <name>WSO2UM_DB</name> <description>The datasource used by user manager</description> <jndiConfig> <name>jdbc/WSO2UM_DB</name> </jndiConfig> <definition type="RDBMS"> <configuration> <url>jdbc:mysql://[MySQL DB url]:[port]/userdb</url> <username>[user]</username> <password>[password]</password> <driverClassName>com.mysql.jdbc.Driver</driverClassName> <maxActive>50</maxActive> <maxWait>60000</maxWait> <testOnBorrow>true</testOnBorrow> <validationQuery>SELECT 1</validationQuery> <validationInterval>30000</validationInterval> </configuration> </definition> </datasource>

      2. Enable the nodes to access the registry database by configuring the WSO2REG_DB data source as follows.

         Code Block
         language xml
         <datasource> <name>WSO2REG_DB</name> <description>The datasource used by the registry</description> <jndiConfig> <name>jdbc/WSO2REG_DB</name> </jndiConfig> <definition type="RDBMS"> <configuration> <url>jdbc:mysql://[MySQL DB url]:[port]/regdb</url> <username>[user]</username> <password>[password]</password> <driverClassName>com.mysql.jdbc.Driver</driverClassName> <maxActive>50</maxActive> <maxWait>60000</maxWait> <testOnBorrow>true</testOnBorrow> <validationQuery>SELECT 1</validationQuery> <validationInterval>30000</validationInterval> </configuration> </definition> </datasource>

         For detailed information about registry sharing strategies, see the library article Sharing Registry Space across Multiple Product Instances.

   2. Point to your database WSO2_ANALYTICS_EVENT_STORE_DB and WSO2_ANALYTICS_PROCESSED_DATA_STORE_DB in the <IS Analytics_HOME>/repository/conf/datasources/analytics-datasources.xml file as shown below.

      Code Block
      language xml
      <datasources-configuration> <providers> <provider>org.wso2.carbon.ndatasource.rdbms.RDBMSDataSourceReader</provider> </providers> <datasources> <datasource> <name>WSO2_ANALYTICS_EVENT_STORE_DB</name> <description>The datasource used for analytics record store</description> <definition type="RDBMS"> <configuration> <url>jdbc:mysql://[MySQL DB url]:[port]/WSO2_ANALYTICS_EVENT_STORE_DB</url> <username>[username]</username> <password>[password]</password> <driverClassName>com.mysql.jdbc.Driver</driverClassName> <maxActive>50</maxActive> <maxWait>60000</maxWait> <testOnBorrow>true</testOnBorrow> <validationQuery>SELECT 1</validationQuery> <validationInterval>30000</validationInterval> <defaultAutoCommit>false</defaultAutoCommit> </configuration> </definition> </datasource> <datasource> <name>WSO2_ANALYTICS_PROCESSED_DATA_STORE_DB</name> <description>The datasource used for analytics record store</description> <definition type="RDBMS"> <configuration> <url>jdbc:mysql://[MySQL DB url]:[port]/WSO2_ANALYTICS_PROCESSED_DATA_STORE_DB</url> <username>[username]</username> <password>[password]</password> <driverClassName>com.mysql.jdbc.Driver</driverClassName> <maxActive>50</maxActive> <maxWait>60000</maxWait> <testOnBorrow>true</testOnBorrow> <validationQuery>SELECT 1</validationQuery> <validationInterval>30000</validationInterval> <defaultAutoCommit>false</defaultAutoCommit> </configuration> </definition> </datasource> </datasources> </datasources-configuration>

      For more information, see Datasources in DAS documentation.

   3. To share the user store among the nodes, open the <IS Analytics_HOME>/repository/conf/user-mgt.xmlfile and modify the dataSource property of the <configuration> element as follows.

      Code Block
      language xml
      <configuration> ... <Property name="dataSource">jdbc/WSO2UM_DB</Property> </configuration>

      Info
      The datasource name specified in this configuration should be the same as the datasource used by user manager that you configured in sub step a, i.

   4. In the <IS Analytics_HOME>/repository/conf/registry.xml file, add or modify the dataSource attribute of the <dbConfig name="govregistry"> element as follows.

      Code Block
      language xml
      <dbConfig name="govregistry"> <dataSource>jdbc/WSO2REG_DB</dataSource> </dbConfig> <remoteInstance url="https://localhost:9443/registry">  <id>gov</id> <cacheId>user@jdbc:mysql://localhost:3306/regdb</cacheId> <dbConfig>govregistry</dbConfig> <readOnly>false</readOnly> <enableCache>true</enableCache> <registryRoot>/</registryRoot> </remoteInstance> <mount path="/_system/governance" overwrite="true"> <instanceId>gov</instanceId> <targetPath>/_system/governance</targetPath> </mount> <mount path="/_system/config" overwrite="true"> <instanceId>gov</instanceId> <targetPath>/_system/config</targetPath> </mount>

      Note
      Do not replace the following configuration when adding in the mounting configurations. The registry mounting configurations mentioned in the above steps should be added in addition to the following. Code Block language xml <dbConfig name="wso2registry"> <dataSource>jdbc/WSO2CarbonDB</dataSource> </dbConfig>

2. Update the <IS Analytics_HOME>/repository/conf/axis2/axis2.xml file as follows to enable Hazlecast clustering for both nodes. 
   

   1. Set clustering class="org.wso2.carbon.core.clustering.hazelcast.HazelcastClusteringAgent" to true as shown below to enable Hazlecast clustering.

      Code Block
      language xml
      <clustering class="org.wso2.carbon.core.clustering.hazelcast.HazelcastClusteringAgent" enable="true">

   2. Enable wka mode on both nodes as shown below.  For more information on wka mode, read About membership schemes.

      Code Block
      language xml
      <parameter name="membershipScheme">wka</parameter>

   3. Add both the nodes as well known members in the cluster under the members tag in each node as shown in the example below.

      Code Block
      language xml
      <members> <member> <hostName>[node1 IP]</hostName> <port>[node1 port]</port> </member> <member> <hostName>[node2 IP]</hostName> <port>[node2 port]</port> </member> </members>

   4. For each node, enter the respective server IP address as the value for the localMemberHost property as shown below.

      Code Block
      language xml
      <parameter name="localMemberHost">[Server_IP_Address]</parameter>

   Expand
   title Click here to view the complete clustering section of the axis2.xml file. with the changes mentioned above.
   Code Block language xml <clustering class="org.wso2.carbon.core.clustering.hazelcast.HazelcastClusteringAgent" enable="true"> <!-- This parameter indicates whether the cluster has to be automatically initalized when the AxisConfiguration is built. If set to "true" the initialization will not be done at that stage, and some other party will have to explictly initialize the cluster. --> <parameter name="AvoidInitiation">true</parameter> <!-- The membership scheme used in this setup. The only values supported at the moment are "multicast" and "wka" 1. multicast - membership is automatically discovered using multicasting 2. wka - Well-Known Address based multicasting. Membership is discovered with the help of one or more nodes running at a Well-Known Address. New members joining a cluster will first connect to a well-known node, register with the well-known node and get the membership list from it. When new members join, one of the well-known nodes will notify the others in the group. When a member leaves the cluster or is deemed to have left the cluster, it will be detected by the Group Membership Service (GMS) using a TCP ping mechanism. --> <parameter name="membershipScheme">wka</parameter> <!--<parameter name="licenseKey">xxx</parameter>--> <!--<parameter name="mgtCenterURL">http://localhost:8081/mancenter/</parameter>--> <!-- The clustering domain/group. Nodes in the same group will belong to the same multicast domain. There will not be interference between nodes in different groups. --> <parameter name="domain">wso2.carbon.domain</parameter> <!-- The multicast address to be used --> <!--<parameter name="mcastAddress">228.0.0.4</parameter>--> <!-- The multicast port to be used --> <parameter name="mcastPort">45564</parameter> <parameter name="mcastTTL">100</parameter> <parameter name="mcastTimeout">60</parameter> <!-- The IP address of the network interface to which the multicasting has to be bound to. Multicasting would be done using this interface. --> <!-- <parameter name="mcastBindAddress">10.100.5.109</parameter> --> <!-- The host name or IP address of this member --> <parameter name="localMemberHost">[node IP]</parameter> <!-- The bind adress of this member. The difference between localMemberHost & localMemberBindAddress is that localMemberHost is the one that is advertised by this member, while localMemberBindAddress is the address to which this member is bound to. --> <!-- <parameter name="localMemberBindAddress">[node IP]</parameter> --> <!-- The TCP port used by this member. This is the port through which other nodes will contact this member --> <parameter name="localMemberPort">[node port]</parameter> <!-- The bind port of this member. The difference between localMemberPort & localMemberBindPort is that localMemberPort is the one that is advertised by this member, while localMemberBindPort is the port to which this member is bound to. --> <!-- <parameter name="localMemberBindPort">4001</parameter> --> <!-- Properties specific to this member --> <parameter name="properties"> <property name="backendServerURL" value="https://${hostName}:${httpsPort}/services/"/> <property name="mgtConsoleURL" value="https://${hostName}:${httpsPort}/"/> <property name="subDomain" value="worker"/> </parameter> <!-- Uncomment the following section to load custom Hazelcast data serializers. --> <!-- <parameter name="hazelcastSerializers"> <serializer typeClass="java.util.TreeSet">org.wso2.carbon.hazelcast.serializer.TreeSetSerializer </serializer> <serializer typeClass="java.util.Map">org.wso2.carbon.hazelcast.serializer.MapSerializer</serializer> </parameter> --> <!-- The list of static or well-known members. These entries will only be valid if the "membershipScheme" above is set to "wka" --> <members> <member> <hostName>[node1 IP]</hostName> <port>[node1 port]</port> </member> <member> <hostName>[node2 IP]</hostName> <port>[node2 port]</port> </member> </members> <!-- Enable the groupManagement entry if you need to run this node as a cluster manager. Multiple application domains with different GroupManagementAgent implementations can be defined in this section. --> <groupManagement enable="false"> <applicationDomain name="wso2.as.domain" description="AS group" agent="org.wso2.carbon.core.clustering.hazelcast.HazelcastGroupManagementAgent" subDomain="worker" port="2222"/> </groupManagement> </clustering>

3. Configure the  <IS Analytics_HOME>/repository/conf/event-processor.xml file as follows to cluster IS Analytics in the Receiver.

   Anchor
   EventProcessor EventProcessor

   1. Enable the HA mode by setting the following property.

      Code Block
      language xml
      <mode name="HA" enable="true">

   2. Disable the Distributed mode by setting the following property.

      Code Block
      language xml
      <mode name="Distributed" enable="false">

   3. For each node, enter the respective server IP address under the HA mode Config section as shown in the example below.

      Info

      When you enable the HA mode for WSO2 IS Analytics, the following are enabled by default: State persistence: If there is no real time use case that requires any state information after starting the cluster, you should disable event persistence by setting the persistence attribute to false in the <IS Analytics_HOME>/repository/conf/event-processor.xml file as shown below. Code Block language xml <persistence enable="false"> <persistenceIntervalInMinutes>15</persistenceIntervalInMinutes> <persisterSchedulerPoolSize>10</persisterSchedulerPoolSize> <persister class="org.wso2.carbon.event.processor.core.internal.persistence.FileSystemPersistenceStore"> <property key="persistenceLocation">cep_persistence</property> </persister> </persistence> Tip When state persistence is enabled for WSO2 IS Analytics, the internal state of IS Analytics is persisted in files. These files are not automatically deleted. Therefore, if you want to save space in your IS Analytics pack, you need to delete them manually. These files are created in the <IS Analytics_HOME>/cep_persistence/<tenant-id> directory. This directory has a separate sub-directory for each execution plan. Each execution plan can have multiple files. The format of each file name is <TIMESTAMP>_<EXECUTION_PLAN_NAME> (e.g, 1493101044948_MyExecutionPlan). If you want to clear files for a specific execution plan, you need to leave the two files with the latest timestamps and delete the rest. Event synchronization: However, if you set the event.duplicated.in.cluster=true property for an event receiver configured in a node, IS Analytics does not perform event synchronization for that receiver.

      Code Block
      language xml
      <!-- HA Mode Config --> <mode name="HA" enable="true"> ... <eventSync> <hostName>[Server_IP_Address]</hostName>

   Expand
   title Click here to view the complete event-processor.xml file with the changes mentioned above.
   Code Block language xml <eventProcessorConfiguration> <mode name="SingleNode" enable="false"> <persistence enable="false"> <persistenceIntervalInMinutes>15</persistenceIntervalInMinutes> <persisterSchedulerPoolSize>10</persisterSchedulerPoolSize> <persister class="org.wso2.carbon.event.processor.core.internal.persistence.FileSystemPersistenceStore"> <property key="persistenceLocation">cep_persistence</property> </persister> </persistence> </mode> <!-- HA Mode Config --> <mode name="HA" enable="true"> <nodeType> <worker enable="true"/> <presenter enable="false"/> </nodeType> <checkMemberUpdateInterval>10000</checkMemberUpdateInterval> <eventSync> <hostName>172.18.1.217</hostName> <port>11224</port> <reconnectionInterval>20000</reconnectionInterval> <serverThreads>20000</serverThreads> <!--Size of TCP event publishing client's send buffer in bytes--> <publisherTcpSendBufferSize>5242880</publisherTcpSendBufferSize> <!--Character encoding of TCP event publishing client--> <publisherCharSet>UTF-8</publisherCharSet> <publisherBufferSize>1024</publisherBufferSize> <publisherConnectionStatusCheckInterval>30000</publisherConnectionStatusCheckInterval> <!--Number of events that could be queued at receiver before they are synced between CEP/DAS nodes--> <receiverQueueSize>1000000</receiverQueueSize> <!--Max total size of events that could be queued at receiver before they are synced between CEP/DAS nodes--> <receiverQueueMaxSizeMb>10</receiverQueueMaxSizeMb> <!--Number of events that could be queued at publisher to sync output between CEP/DAS nodes--> <publisherQueueSize>1000000</publisherQueueSize> <!--Max total size of events that could be queued at publisher to sync output between CEP/DAS nodes--> <publisherQueueMaxSizeMb>10</publisherQueueMaxSizeMb> </eventSync> <management> <hostName>172.18.1.217</hostName> <port>10005</port> <tryStateChangeInterval>15000</tryStateChangeInterval> <stateSyncRetryInterval>10000</stateSyncRetryInterval> </management> <presentation> <hostName>0.0.0.0</hostName> <port>11000</port> <!--Size of TCP event publishing client's send buffer in bytes--> <publisherTcpSendBufferSize>5242880</publisherTcpSendBufferSize> <!--Character encoding of TCP event publishing client--> <publisherCharSet>UTF-8</publisherCharSet> <publisherBufferSize>1024</publisherBufferSize> <publisherConnectionStatusCheckInterval>30000</publisherConnectionStatusCheckInterval> </presentation> </mode> <!-- Distributed Mode Config --> <mode name="Distributed" enable="false"> <nodeType> <worker enable="true"/> <manager enable="true"> <hostName>0.0.0.0</hostName> <port>8904</port> </manager> <presenter enable="false"> <hostName>0.0.0.0</hostName> <port>11000</port> </presenter> </nodeType> <management> <managers> <manager> <hostName>localhost</hostName> <port>8904</port> </manager> <manager> <hostName>localhost</hostName> <port>8905</port> </manager> </managers> <!--Connection re-try interval to connect to Storm Manager service in case of a connection failure--> <reconnectionInterval>20000</reconnectionInterval> <!--Heart beat interval (in ms) for event listeners in "Storm Receivers" and "CEP Publishers" to acknowledge their availability for receiving events"--> <heartbeatInterval>5000</heartbeatInterval> <!--Storm topology re-submit interval in case of a topology submission failure--> <topologyResubmitInterval>10000</topologyResubmitInterval> </management> <transport> <!--Port range to be used for events listener servers in "Storm Receiver Spouts" and "CEP Publishers"--> <portRange> <min>15000</min> <max>15100</max> </portRange> <!--Connection re-try interval (in ms) for connection failures between "CEP Receiver" to "Storm Receiver" connections and "Storm Publisher" to "CEP Publisher" connections--> <reconnectionInterval>20000</reconnectionInterval> <!--Size of the output queue of each "CEP Receiver" which stores events to be published into "Storm Receivers" . This must be a power of two--> <cepReceiverOutputQueueSize>8192</cepReceiverOutputQueueSize> <!--Size of the output queue of each "Storm Publisher" which stores events to be published into "CEP Publisher" . This must be a power of two--> <stormPublisherOutputQueueSize>8192</stormPublisherOutputQueueSize> <!--Size of TCP event publishing client's send buffer in bytes--> <tcpEventPublisherSendBufferSize>5242880</tcpEventPublisherSendBufferSize> <!--Character encoding of TCP event publishing client--> <tcpEventPublisherCharSet>UTF-8</tcpEventPublisherCharSet> <!--Size of the event queue in each storm spout which stores events to be processed by storm bolts --> <stormSpoutBufferSize>10000</stormSpoutBufferSize> <connectionStatusCheckInterval>20000</connectionStatusCheckInterval> </transport> <presentation> <presentationOutputQueueSize>1024</presentationOutputQueueSize> <!--Size of TCP event publishing client's send buffer in bytes--> <tcpEventPublisherSendBufferSize>5242880</tcpEventPublisherSendBufferSize> <!--Character encoding of TCP event publishing client--> <tcpEventPublisherCharSet>UTF-8</tcpEventPublisherCharSet> <connectionStatusCheckInterval>20000</connectionStatusCheckInterval> </presentation> <statusMonitor> <lockTimeout>60000</lockTimeout> <updateRate>60000</updateRate> </statusMonitor> <stormJar>org.wso2.cep.storm.dependencies.jar</stormJar> <distributedUIUrl></distributedUIUrl> <memberUpdateCheckInterval>20000</memberUpdateCheckInterval> </mode> </eventProcessorConfiguration>

   Info

   The following node types are configured for the HA deployment mode in the <IS Analytics_HOME>/repository/conf/event-processor.xml file. eventSync: Both the active and the passive nodes in this setup are event synchronizing nodes as explained in the introduction. Therefore, each node should have the host and the port on which it is operating specified under the <eventSync> element. Info Note that the eventSync port is not automatically updated to the port in which each node operates via port offset. management: In this setup, both the nodes carry out the same tasks, and therefore, both nodes are considered manager nodes. Therefore, each node should have the host and the port on which it is operating specified under the <management> element. Info Note that the management port is not automatically updated to the port in which each node operates via port offset. presentation: You can optionally specify only one of the two nodes in this setup as the presenter node. The dashboards in which processed information is displayed are configured only in the presenter node. Each node should have the host and the port on which the assigned presenter node is operating specified under the <presentation> element. The host and the port as well as the other configurations under the <presentation> element are effective only when the presenter enable="false property is set under the <!-- HA Mode Config --> section.

4. Update the <IS Analytics_HOME>/repository/conf/analytics/spark/spark-defaults.conf file as follows to use the Spark cluster embedded within IS Analytics.

   • Keep the carbon.spark.master configuration as local. This instructs Spark to create a Spark cluster using the Hazelcast cluster.

   • Enter 2 as the value for the carbon.spark.master.count configuration. This specifies that there should be two masters controllers in the Spark cluster. One master controller serves as an active master controller and the other serves as a stand-by mastercontroller. 

   The following example shows the <IS Analytics_HOME>/repository/conf/analytics/spark/spark-defaults.conf file with changes mentioned above.

   Code Block
   carbon.spark.master local carbon.spark.master.count 2

   For more information, see Spark Configurations in DAS documentation.

   Warning

   Important: If the path to <IS Analytics_HOME> is different in the two nodes, please do the following.  Localtabgroup Localtab active true title UNIX environment Create a symbolic link to <IS Analytics_HOME> in both nodes, where paths of those symbolic links are identical. This ensures that if we use the symbolic link to access IS Analytics, we can use a common path. To do this, set the following property in the <IS Analytics_HOME>/repository/conf/analytics/spark/spark-defaults.conf file. carbon.das.symbolic.link /home/ubuntu/das/das_symlink/ Localtab title Windows environment In the Windows environment there is a strict requirement to have both IS Analytics distributions in a common path.

5. In order to share the C-Apps deployed among the nodes, configure the SVN-based deployment synchronizer. For detailed instructions, see Configuring SVN-Based Deployment Synchronizer.

   Tip
   IS Analytics Minimum High availability Deployment set up does not use a manager and a worker. For the purpose of configuring the deployment synchronizer, you can add the configurations relevant to the manager for the node of your choice, and add the configurations relating to the worker for the other node.

   Info
   If you do not configure the deployment synchronizer, you are required to deploy any C-App you use in the IS Analytics Minimum High Availability Deployment set up to both the nodes.

6. If the physical IS Analytics server has multiple network interfaces with different IPs, and if you want Spark to use a specific Interface IP, open either the <IS Analytics_HOME>/bin/load-spark-env-vars.sh file (for Linux) or <IS Analytics_HOME>/bin/load-spark-env-vars.bat file (for Windows), and add the following parameter to configure the Spark IP address. 

   Code Block
   export SPARK_LOCAL_IP=<IP_Address>

...

1. Access the Spark UIs of the active master and controller and the stand-by master controller using < node ip>:8081 in each node.
   • Information relating to the active master controller is displayed as shown in the example below.
     
   • Information relating to the stand-by master is controller is displayed as shown in the example below.
     
2. Click the links under Running Applications in the Spark UI of the active master to controller to check the Spark application UIs of those applications. A working application is displayed as shown in the following example.
   
3. Click the Environment tab of a Spark application UI to check whether all the configuration parameters are correctly set. You can also check whether the class path variables in this tab can be accessed manually.
   
4. Check the Spark UIs of workers to check whether they have running executors. If a worker UI does not have running executors or if it is continuously creating executors, it indicates an issue in the Spark cluster configuration. The following example shows a worker UI with a running executor.
   
5. Check the symbolic parameter, and check if you could manually access it via a cd <directory> command in the CLI.
6. Log into the IS Analytics Management Console and navigate to Main => Manage => Batch Analytics => Console to open the Interactive Analytics Console. Run a query in this console.