hanavora

SAP Vora 1.2 - Reading Vora tables from HANA


!!! UPDATE !!!

Finally after hours of looking into documentation I found the issue. It turns out that I lacked some parameters in Yarn configuration.

This is what I did:

  1. Open the yarn-site.xml file in an editor or log in to Ambari web UI and select Yarn>Config. Locate the property "yarn.nodemanager.aux-services" and add "spark_shuffle" to its current value. The new property name should be "mapreduce_shuffle,spark_shuffle".
  2. Add or edit the property "yarn.nodemanager.aux-services.spark_shuffle.class", and set it to "org.apache.spark.network.yarn.YarnShuffleService".
  3. Copy the spark--yarn-shuffle.jar file (downloaded in the step Install Spark Assembly Files and Dependent Libraries) from Spark to Hadoop-Yarn class path in all the node manager hosts. Typically this folder is located in /usr/hdp//hadoop-yarn/lib.
  4. Restart Yarn and the node manager

!!!!!!!!!!!

I'm using SAP Vora 1.2 Developer Edition with newest Spark Controller (HANASPARKCTRL00P_5-70001262.RPM). I loaded a table into Vora in spark-shell. I can see the table in SAP HANA Studio in "spark_velocity" folder. I can load the table as Virtual Table. The problem is that I cannot select, or preview the data in the table, because of the error:

Error: SAP DBTech JDBC: [403]: internal error: Error opening the cursor for the remote database for query "SELECT "SPARK_testtable"."a1", "SPARK_testtable"."a2", "SPARK_testtable"."a3" FROM "spark_velocity"."testtable" "SPARK_testtable" LIMIT 200 "

Here is my hanaes-site.xml file:

<configuration>
    <!--  You can either copy the assembly jar into HDFS or to lib/external directory.
    Please maintain appropriate value here-->
    <property>
        <name>sap.hana.es.spark.yarn.jar</name>
        <value>file:///usr/sap/spark/controller/lib/external/spark-assembly-1.5.2.2.3.4.0-3485-hadoop2.7.1.2.3.4.0-3485.jar</value>
        <final>true</final>
    </property>
    <property>
        <name>sap.hana.es.server.port</name>
        <value>7860</value>
        <final>true</final>
    </property>
    <!--  Required if you are copying your files into HDFS-->
     <property>
         <name>sap.hana.es.lib.location</name>
         <value>hdfs:///sap/hana/spark/libs/thirdparty/</value>
         <final>true</final>
     </property>
     -->
    <!--Required property if using controller for DLM scenarios-->
    <!--
    <property>
        <name>sap.hana.es.warehouse.dir</name>
        <value>/sap/hana/hanaes/warehouse</value>
        <final>true</final>
    </property>
-->
    <property>
        <name>sap.hana.es.driver.host</name>
        <value>ip-10-0-0-[censored].ec2.internal</value>
        <final>true</final>
    </property>
    <!-- Change this value to vora when connecting to Vora store -->
    <property>
        <name>sap.hana.hadoop.datastore</name>
        <value>vora</value>
        <final>true</final>
    </property>

    <!-- // When running against a kerberos protected cluster, please maintain appropriate values
    <property>
        <name>spark.yarn.keytab</name>
        <value>/usr/sap/spark/controller/conf/hanaes.keytab</value>
        <final>true</final>
    </property>
    <property>
        <name>spark.yarn.principal</name>
        <value>hanaes@PAL.SAP.CORP</value>
        <final>true</final>
    </property>
-->
    <!-- To enable Secure Socket communication, please maintain appropriate values in the follwing section-->
    <property>
        <name>sap.hana.es.ssl.keystore</name>
        <value></value>
        <final>false</final>
    </property>
    <property>
        <name>sap.hana.es.ssl.clientauth.required</name>
        <value>true</value>
        <final>true</final>
    </property>
    <property>
        <name>sap.hana.es.ssl.verify.hostname</name>
        <value>true</value>
        <final>true</final>
    </property>
    <property>
        <name>sap.hana.es.ssl.keystore.password</name>
        <value></value>
        <final>true</final>
    </property>
    <property>
        <name>sap.hana.es.ssl.truststore</name>
        <value></value>
        <final>true</final>
    </property>
    <property>
        <name>sap.hana.es.ssl.truststore.password</name>
        <value></value>
        <final>true</final>
    </property>
    <property>
        <name>sap.hana.es.ssl.enabled</name>
        <value>false</value>
        <final>true</final>
    </property>

    <property>
        <name>spark.executor.instances</name>
        <value>10</value>
        <final>true</final>
    </property>
    <property>
        <name>spark.executor.memory</name>
        <value>5g</value>
        <final>true</final>
    </property>
    <!-- Enable the following section if you want to enable dynamic allocation-->
    <!--
    <property>
        <name>spark.dynamicAllocation.enabled</name>
        <value>true</value>
        <final>true</final>
    </property>

    <property>
        <name>spark.dynamicAllocation.minExecutors</name>
        <value>10</value>
        <final>true</final>
    </property>
    <property>
        <name>spark.dynamicAllocation.maxExecutors</name>
        <value>20</value>
        <final>true</final>
    </property>
    <property>
    <name>spark.shuffle.service.enabled</name>
    <value>true</value>
    <final>true</final>
   </property>
<property>
         <name>sap.hana.ar.provider</name>
         <value>com.sap.hana.aws.extensions.AWSResolver</value>
         <final>true</final>
     </property>
<property>
        <name>spark.vora.hosts</name>
        <value>ip-10-0-0-[censored].ec2.internal:2022,ip-10-0-0-[censored].ec2.internal:2022,ip-10-0-0-[censored].ec2.internal:2022</value>
        <final>true</final>
     </property>
     <property>
        <name>spark.vora.zkurls</name>
        <value>ip-10-0-0-[censored].ec2.internal:2181,ip-10-0-0-[censored].ec2.internal:2181,ip-10-0-0-[censored].ec2.internal:2181</value>
        <final>true</final>
     </property>
</configuration>

ls /usr/sap/spark/controller/lib/external/

spark-assembly-1.5.2.2.3.4.0-3485-hadoop2.7.1.2.3.4.0-3485.jar

hdfs dfs -ls /sap/hana/spark/libs/thirdparty

Found 4 items
-rwxrwxrwx   3 hdfs hdfs     366565 2016-05-11 13:09 /sap/hana/spark/libs/thirdparty/datanucleus-api-jdo-4.2.1.jar
-rwxrwxrwx   3 hdfs hdfs    2006182 2016-05-11 13:09 /sap/hana/spark/libs/thirdparty/datanucleus-core-4.1.2.jar
-rwxrwxrwx   3 hdfs hdfs    1863315 2016-05-11 13:09 /sap/hana/spark/libs/thirdparty/datanucleus-rdbms-4.1.2.jar
-rwxrwxrwx   3 hdfs hdfs     627814 2016-05-11 13:09 /sap/hana/spark/libs/thirdparty/joda-time-2.9.3.jar

ls /usr/hdp/

2.3.4.0-3485  2.3.4.7-4  current

vi /var/log/hanaes/hana_controller.log

SLF4J: Class path contains multiple SLF4J bindings.
SLF4J: Found binding in [jar:file:/usr/sap/spark/controller/lib/spark-sap-datasources-1.2.33-assembly.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: Found binding in [jar:file:/usr/sap/spark/controller/lib/external/spark-assembly-1.5.2.2.3.4.0-3485-hadoop2.7.1.2.3.4.0-3485.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: Found binding in [jar:file:/usr/hdp/2.3.4.0-3485/hadoop/lib/slf4j-log4j12-1.7.10.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory]
16/05/12 07:02:38 INFO HanaESConfig: Loaded HANA Extended Store Configuration
Found Spark Libraries. Proceeding with Current Class Path
16/05/12 07:02:39 INFO Server: Starting Spark Controller
16/05/12 07:03:11 INFO CommandRouter: Connecting to Vora Engine
16/05/12 07:03:11 INFO CommandRouter: Initialized Router
16/05/12 07:03:11 INFO CommandRouter: Server started
16/05/12 07:03:43 INFO CommandHandler: Getting BROWSE data/user/17401406272892502037-4985062628452729323_f17e36cf-0003-0015-452e-800c700001ee
16/05/12 07:03:48 INFO CommandHandler: Getting BROWSE data/user/17401406272892502037-4985062628452729329_f17e36cf-0003-0015-452e-800c700001f4
16/05/12 07:03:48 INFO VoraClientFactory: returning a Vora catalog client of this Vora catalog server: master.i-14371789.cluster:2204
16/05/12 07:03:48 INFO CBinder: searching for compat-sap-c++.so at /opt/rh/SAP/lib64/compat-sap-c++.so
16/05/12 07:03:48 WARN CBinder: could not find compat-sap-c++.so
16/05/12 07:03:48 INFO CBinder: searching for libpam.so.0 at /lib64/libpam.so.0
16/05/12 07:03:48 INFO CBinder: loading libpam.so.0 from /lib64/libpam.so.0
16/05/12 07:03:48 INFO CBinder: loading library libprotobuf.so
16/05/12 07:03:48 INFO CBinder: loading library libprotoc.so
16/05/12 07:03:48 INFO CBinder: loading library libtbbmalloc.so
16/05/12 07:03:48 INFO CBinder: loading library libtbb.so
16/05/12 07:03:48 INFO CBinder: loading library libv2runtime.so
16/05/12 07:03:48 INFO CBinder: loading library libv2net.so
16/05/12 07:03:48 INFO CBinder: loading library libv2catalog_connector.so
16/05/12 07:03:48 INFO CatalogFactory: returning a Vora catalog client of this Vora catalog server: master.i-14371789.cluster:2204
16/05/12 07:11:56 INFO CommandHandler: Getting BROWSE data/user/17401406272892502037-4985062628452729335_f17e36cf-0003-0015-452e-800c700001fa
16/05/12 07:11:56 INFO Utils: freeing the buffer
16/05/12 07:11:56 INFO Utils: freeing the buffer
16/05/12 07:12:02 INFO Utils: freeing the buffer
16/05/12 07:12:02 WARN DefaultSource: Creating a Vora Relation that is actually persistent with a temporary statement!
16/05/12 07:12:02 WARN DefaultSource: Creating a Vora Relation that is actually persistent with a temporary statement!
16/05/12 07:12:02 INFO CatalogFactory: returning a Vora catalog client of this Vora catalog server: master.i-14371789.cluster:2204
16/05/12 07:12:02 INFO Utils: freeing the buffer
16/05/12 07:12:02 INFO DefaultSource: Creating VoraRelation testtable using an existing catalog table
16/05/12 07:12:02 INFO Utils: freeing the buffer
16/05/12 07:12:11 INFO Utils: freeing the buffer
16/05/12 07:14:15 ERROR RequestOrchestrator: Result set was not fetched by connected Client. Hence cancelled the execution
16/05/12 07:14:15 ERROR RequestOrchestrator: org.apache.spark.SparkException: Job 0 cancelled part of cancelled job group f17e36cf-0003-0015-452e-800c70000216
        at org.apache.spark.scheduler.DAGScheduler.org$apache$spark$scheduler$DAGScheduler$$failJobAndIndependentStages(DAGScheduler.scala:1283)
        at org.apache.spark.scheduler.DAGScheduler.handleJobCancellation(DAGScheduler.scala:1229)
        at org.apache.spark.scheduler.DAGScheduler$$anonfun$handleJobGroupCancelled$1.apply$mcVI$sp(DAGScheduler.scala:681)
        at org.apache.spark.scheduler.DAGScheduler$$anonfun$handleJobGroupCancelled$1.apply(DAGScheduler.scala:681)
        at org.apache.spark.scheduler.DAGScheduler$$anonfun$handleJobGroupCancelled$1.apply(DAGScheduler.scala:681)
        at scala.collection.mutable.HashSet.foreach(HashSet.scala:79)
        at org.apache.spark.scheduler.DAGScheduler.handleJobGroupCancelled(DAGScheduler.scala:681)
        at org.apache.spark.scheduler.DAGSchedulerEventProcessLoop.doOnReceive(DAGScheduler.scala:1475)
        at org.apache.spark.scheduler.DAGSchedulerEventProcessLoop.onReceive(DAGScheduler.scala:1458)
        at org.apache.spark.scheduler.DAGSchedulerEventProcessLoop.onReceive(DAGScheduler.scala:1447)
        at org.apache.spark.util.EventLoop$$anon$1.run(EventLoop.scala:48)
        at org.apache.spark.scheduler.DAGScheduler.runJob(DAGScheduler.scala:567)
        at org.apache.spark.SparkContext.runJob(SparkContext.scala:1824)
        at org.apache.spark.SparkContext.runJob(SparkContext.scala:1837)
        at org.apache.spark.SparkContext.runJob(SparkContext.scala:1850)
        at org.apache.spark.SparkContext.runJob(SparkContext.scala:1921)
        at org.apache.spark.rdd.RDD$$anonfun$foreachPartition$1.apply(RDD.scala:902)
        at org.apache.spark.rdd.RDD$$anonfun$foreachPartition$1.apply(RDD.scala:900)
        at org.apache.spark.rdd.RDDOperationScope$.withScope(RDDOperationScope.scala:147)
        at org.apache.spark.rdd.RDDOperationScope$.withScope(RDDOperationScope.scala:108)
        at org.apache.spark.rdd.RDD.withScope(RDD.scala:310)
        at org.apache.spark.rdd.RDD.foreachPartition(RDD.scala:900)
        at com.sap.hana.spark.network.CommandHandler$$anonfun$receive$2$$anonfun$applyOrElse$7.apply(CommandRouter.scala:383)
        at com.sap.hana.spark.network.CommandHandler$$anonfun$receive$2$$anonfun$applyOrElse$7.apply(CommandRouter.scala:362)
        at scala.collection.immutable.List.foreach(List.scala:318)
        at com.sap.hana.spark.network.CommandHandler$$anonfun$receive$2.applyOrElse(CommandRouter.scala:362)
        at akka.actor.Actor$class.aroundReceive(Actor.scala:467)
        at com.sap.hana.spark.network.CommandHandler.aroundReceive(CommandRouter.scala:204)
        at akka.actor.ActorCell.receiveMessage(ActorCell.scala:516)
        at akka.actor.ActorCell.invoke(ActorCell.scala:487)
        at akka.dispatch.Mailbox.processMailbox(Mailbox.scala:238)
        at akka.dispatch.Mailbox.run(Mailbox.scala:220)
        at akka.dispatch.ForkJoinExecutorConfigurator$AkkaForkJoinTask.exec(AbstractDispatcher.scala:397)
        at scala.concurrent.forkjoin.ForkJoinTask.doExec(ForkJoinTask.java:260)
        at scala.concurrent.forkjoin.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:1339)
        at scala.concurrent.forkjoin.ForkJoinPool.runWorker(ForkJoinPool.java:1979)
        at scala.concurrent.forkjoin.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:107)

Also strange is this error:

16/05/12 07:03:48 INFO CBinder: searching for compat-sap-c++.so at /opt/rh/SAP/lib64/compat-sap-c++.so
    16/05/12 07:03:48 WARN CBinder: could not find compat-sap-c++.so

Because I have this file in the location:

ls /opt/rh/SAP/lib64/

compat-sap-c++.so

After changing com.sap.hana.aws.extensions.AWSResolver into com.sap.hana.spark.aws.extensions.AWSResolver now the log file looks different:

    16/05/17 10:04:08 INFO CommandHandler: Getting BROWSE data/user/9110494231822270485-5373255807276155190_7e6efa3c-0003-0015-4a91-a3b020000139
16/05/17 10:04:13 INFO CommandHandler: Getting BROWSE data/user/9110494231822270485-5373255807276155196_7e6efa3c-0003-0015-4a91-a3b02000013f
16/05/17 10:04:13 INFO Utils: freeing the buffer
16/05/17 10:04:13 INFO Utils: freeing the buffer
16/05/17 10:04:13 INFO Utils: freeing the buffer
16/05/17 10:04:13 INFO Utils: freeing the buffer
16/05/17 10:04:29 INFO Utils: freeing the buffer
16/05/17 10:04:29 WARN DefaultSource: Creating a Vora Relation that is actually persistent with a temporary statement!
16/05/17 10:04:29 WARN DefaultSource: Creating a Vora Relation that is actually persistent with a temporary statement!
16/05/17 10:04:29 INFO Utils: freeing the buffer
16/05/17 10:04:29 INFO DefaultSource: Creating VoraRelation testtable using an existing catalog table
16/05/17 10:04:29 INFO Utils: freeing the buffer
16/05/17 10:04:29 INFO Utils: freeing the buffer
16/05/17 10:04:29 INFO Utils: freeing the buffer
16/05/17 10:04:29 INFO ConfigurableHostMapper: Load Strategy: RELAXEDLOCAL (default)
16/05/17 10:04:29 INFO HdfsBlockRetriever: Length of HDFS file (/user/vora/test.csv): 10 bytes.
16/05/17 10:04:29 INFO Utils: freeing the buffer
16/05/17 10:04:29 INFO ConfigurableHostMapper: Load Strategy: RELAXEDLOCAL (default)
16/05/17 10:04:29 INFO TableLoader: Loading table [testtable]
16/05/17 10:04:29 INFO ConfigurableHostMapper: Load Strategy: RELAXEDLOCAL (default)
16/05/17 10:04:29 INFO TableLoader: Initialized 1 loading threads. Waiting until finished... -- 0.00 s
16/05/17 10:04:29 INFO TableLoader: [secondary2.i-a5361638.cluster:2202] Host mapping (Ranges: 1/1 Size: 0.00 MB)
16/05/17 10:04:29 INFO VoraJdbcClient: [secondary2.i-a5361638.cluster:2202] MultiLoad: MULTIFILE
16/05/17 10:04:29 INFO TableLoader: [secondary2.i-a5361638.cluster:2202] Host finished:
    Raw ranges: 1/1
    Size:       0.00 MB
    Time:       0.29 s
    Throughput: 0.00 MB/s
16/05/17 10:04:29 INFO TableLoader: Finished 1 loading threads. -- 0.29 s
16/05/17 10:04:29 INFO TableLoader: Updated catalog -- 0.01 s
16/05/17 10:04:29 INFO TableLoader: Table load statistics:
    Name: testtable
    Size: 0.00 MB
    Hosts: 1
    Time: 0.30 s
    Cluster throughput: 0.00 MB/s
    Avg throughput per host: 0.00 MB/s
16/05/17 10:04:29 INFO Utils: freeing the buffer
16/05/17 10:04:29 INFO TableLoader: Loaded table [testtable] -- 0.37 s
16/05/17 10:04:38 INFO Utils: freeing the buffer
16/05/17 10:06:43 ERROR RequestOrchestrator: Result set was not fetched by connected Client. Hence cancelled the execution
16/05/17 10:06:43 ERROR RequestOrchestrator: org.apache.spark.SparkException: Job 1 cancelled part of cancelled job group 7e6efa3c-0003-0015-4a91-a3b02000015b
        at org.apache.spark.scheduler.DAGScheduler.org$apache$spark$scheduler$DAGScheduler$$failJobAndIndependentStages(DAGScheduler.scala:1283)
        at org.apache.spark.scheduler.DAGScheduler.handleJobCancellation(DAGScheduler.scala:1229)
        at org.apache.spark.scheduler.DAGScheduler$$anonfun$handleJobGroupCancelled$1.apply$mcVI$sp(DAGScheduler.scala:681)
        at org.apache.spark.scheduler.DAGScheduler$$anonfun$handleJobGroupCancelled$1.apply(DAGScheduler.scala:681)
        at org.apache.spark.scheduler.DAGScheduler$$anonfun$handleJobGroupCancelled$1.apply(DAGScheduler.scala:681)
        at scala.collection.mutable.HashSet.foreach(HashSet.scala:79)
        at org.apache.spark.scheduler.DAGScheduler.handleJobGroupCancelled(DAGScheduler.scala:681)
        at org.apache.spark.scheduler.DAGSchedulerEventProcessLoop.doOnReceive(DAGScheduler.scala:1475)
        at org.apache.spark.scheduler.DAGSchedulerEventProcessLoop.onReceive(DAGScheduler.scala:1458)
        at org.apache.spark.scheduler.DAGSchedulerEventProcessLoop.onReceive(DAGScheduler.scala:1447)
        at org.apache.spark.util.EventLoop$$anon$1.run(EventLoop.scala:48)
        at org.apache.spark.scheduler.DAGScheduler.runJob(DAGScheduler.scala:567)
        at org.apache.spark.SparkContext.runJob(SparkContext.scala:1824)
        at org.apache.spark.SparkContext.runJob(SparkContext.scala:1837)
        at org.apache.spark.SparkContext.runJob(SparkContext.scala:1850)
        at org.apache.spark.SparkContext.runJob(SparkContext.scala:1921)
        at org.apache.spark.rdd.RDD$$anonfun$foreachPartition$1.apply(RDD.scala:902)
        at org.apache.spark.rdd.RDD$$anonfun$foreachPartition$1.apply(RDD.scala:900)
        at org.apache.spark.rdd.RDDOperationScope$.withScope(RDDOperationScope.scala:147)
        at org.apache.spark.rdd.RDDOperationScope$.withScope(RDDOperationScope.scala:108)
        at org.apache.spark.rdd.RDD.withScope(RDD.scala:310)
        at org.apache.spark.rdd.RDD.foreachPartition(RDD.scala:900)
        at com.sap.hana.spark.network.CommandHandler$$anonfun$receive$2$$anonfun$applyOrElse$7.apply(CommandRouter.scala:383)
        at com.sap.hana.spark.network.CommandHandler$$anonfun$receive$2$$anonfun$applyOrElse$7.apply(CommandRouter.scala:362)
        at scala.collection.immutable.List.foreach(List.scala:318)
        at com.sap.hana.spark.network.CommandHandler$$anonfun$receive$2.applyOrElse(CommandRouter.scala:362)
        at akka.actor.Actor$class.aroundReceive(Actor.scala:467)
        at com.sap.hana.spark.network.CommandHandler.aroundReceive(CommandRouter.scala:204)
        at akka.actor.ActorCell.receiveMessage(ActorCell.scala:516)
        at akka.actor.ActorCell.invoke(ActorCell.scala:487)
        at akka.dispatch.Mailbox.processMailbox(Mailbox.scala:238)
        at akka.dispatch.Mailbox.run(Mailbox.scala:220)
        at akka.dispatch.ForkJoinExecutorConfigurator$AkkaForkJoinTask.exec(AbstractDispatcher.scala:397)
        at scala.concurrent.forkjoin.ForkJoinTask.doExec(ForkJoinTask.java:260)
        at scala.concurrent.forkjoin.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:1339)
        at scala.concurrent.forkjoin.ForkJoinPool.runWorker(ForkJoinPool.java:1979)
        at scala.concurrent.forkjoin.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:107)

I is still "not fetched by the client", but now it looks that vora loaded the table.

Anyone, some ideas how to fix it? The same error appears when I try to read Hive tables insted of Vora.

Error: SAP DBTech JDBC: [403]: internal error: Error opening the cursor for the remote database for query "SELECT "vora_conn_testtable"."a1", "vora_conn_testtable"."a2", "vora_conn_testtable"."a3" FROM "spark_velocity"."testtable" "vora_conn_testtable" LIMIT 200 "


Solution

  • Finally after hours of looking into documentation I found the issue. It turns out that I lacked some parameters in Yarn configuration (don't know why this affected HANA-Vora connection).

    This is what I did:

    Open the yarn-site.xml file in an editor or log in to Ambari web UI and select Yarn>Config. Locate the property "yarn.nodemanager.aux-services" and add "spark_shuffle" to its current value. The new property name should be "mapreduce_shuffle,spark_shuffle". Add or edit the property "yarn.nodemanager.aux-services.spark_shuffle.class", and set it to "org.apache.spark.network.yarn.YarnShuffleService". Copy the spark--yarn-shuffle.jar file from Spark to Hadoop-Yarn class path in all the node manager hosts. Typically this folder is located in /usr/hdp//hadoop-yarn/lib. Restart Yarn and the node manager