dnet-hadoop/dhp-workflows/dhp-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/monitor/oozie_app/workflow.xml

<workflow-app name="Monitor DB" xmlns="uri:oozie:workflow:0.5">
    <parameters>
        <property>
            <name>stats_db_name</name>
            <description>the target stats database name</description>
        </property>
        <property>
            <name>stats_db_shadow_name</name>
            <description>the name of the shadow schema</description>
        </property>
        <property>
            <name>monitor_db_name</name>
            <description>the target monitor db name</description>
        </property>
        <property>
            <name>monitor_db_shadow_name</name>
            <description>the name of the shadow monitor db</description>
        </property>
        <property>
            <name>stats_tool_api_url</name>
            <description>The url of the API of the stats tool. Is used to trigger the cache update.</description>
        </property>
        <property>
            <name>hive_metastore_uris</name>
            <description>hive server metastore URIs</description>
        </property>
        <property>
            <name>hive_jdbc_url</name>
            <description>hive server jdbc url</description>
        </property>
        <property>
            <name>hive_timeout</name>
            <description>the time period, in seconds, after which Hive fails a transaction if a Hive client has not sent a hearbeat. The default value is 300 seconds.</description>
        </property>
        <property>
            <name>context_api_url</name>
            <description>the base url of the context api (https://services.openaire.eu/openaire)</description>
        </property>
    </parameters>

    <global>
        <job-tracker>${jobTracker}</job-tracker>
        <name-node>${nameNode}</name-node>
        <configuration>
            <property>
                <name>hive.metastore.uris</name>
                <value>${hive_metastore_uris}</value>
            </property>
            <property>
            	<name>hive.txn.timeout</name>
            	<value>${hive_timeout}</value>
            </property>
        </configuration>
    </global>

    <start to="Step1-createMonitorDB"/>

    <kill name="Kill">
        <message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
    </kill>

    <action name="Step1-createMonitorDB">
        <shell xmlns="uri:oozie:shell-action:0.1">
            <job-tracker>${jobTracker}</job-tracker>
            <name-node>${nameNode}</name-node>
            <exec>monitor.sh</exec>
            <argument>${stats_db_name}</argument>
            <argument>${monitor_db_name}</argument>
            <argument>${wf:appPath()}/scripts/createMonitorDB.sql</argument>
            <file>monitor.sh</file>
        </shell>
        <ok to="Step2-createMonitorDB-post"/>
        <error to="Kill"/>
    </action>


    <action name="Step2-createMonitorDB-post">
        <shell xmlns="uri:oozie:shell-action:0.1">
            <job-tracker>${jobTracker}</job-tracker>
            <name-node>${nameNode}</name-node>
            <exec>monitor-post.sh</exec>
            <argument>${stats_db_name}</argument>
            <argument>${monitor_db_name}</argument>
            <argument>${monitor_db_shadow_name}</argument>
            <file>monitor-post.sh</file>
        </shell>
        <ok to="Step3-copyDataToImpalaCluster"/>
        <error to="Kill"/>
    </action>

    <action name="Step3-copyDataToImpalaCluster">
        <shell xmlns="uri:oozie:shell-action:0.1">
            <job-tracker>${jobTracker}</job-tracker>
            <name-node>${nameNode}</name-node>
            <exec>copyDataToImpalaCluster.sh</exec>
            <argument>${monitor_db_name}</argument>
            <argument>${hadoop_user_name}</argument>
            <file>copyDataToImpalaCluster.sh</file>
        </shell>
        <ok to="Step4-finalizeImpalaCluster"/>
        <error to="Kill"/>
    </action>

    <action name="Step4-finalizeImpalaCluster">
        <shell xmlns="uri:oozie:shell-action:0.1">
            <job-tracker>${jobTracker}</job-tracker>
            <name-node>${nameNode}</name-node>
            <exec>finalizeImpalaCluster.sh</exec>
            <argument>${monitor_db_name}</argument>
            <argument>${monitor_db_shadow_name}</argument>
            <file>finalizeImpalaCluster.sh</file>
        </shell>
        <ok to="Step5-updateCache"/>
        <error to="Kill"/>
    </action>

    <action name="Step5-updateCache">
        <shell xmlns="uri:oozie:shell-action:0.1">
            <job-tracker>${jobTracker}</job-tracker>
            <name-node>${nameNode}</name-node>
            <exec>updateCache.sh</exec>
            <argument>${stats_tool_api_url}</argument>
            <file>updateCache.sh</file>
        </shell>
        <ok to="End"/>
        <error to="Kill"/>
    </action>
    <end name="End"/>
</workflow-app>