[jira] [Commented] (KYLIN-3137) Update hive-site.xml locating logic for CDH distro
[
https://issues.apache.org/jira/browse/KYLIN-3137?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16497653#comment-16497653
]
ASF subversion and git services commented on KYLIN-3137:
Commit e0f29e29164d44581c133314ac01c419e4277c35 in kylin's branch
refs/heads/master from shaofengshi
[ https://gitbox.apache.org/repos/asf?p=kylin.git;h=e0f29e2 ]
KYLIN-3137 Spark cubing refine
> Update hive-site.xml locating logic for CDH distro
> --
>
> Key: KYLIN-3137
> URL: https://issues.apache.org/jira/browse/KYLIN-3137
> Project: Kylin
> Issue Type: Improvement
> Components: Job Engine, Others, Spark Engine
>Affects Versions: v2.2.0
>Reporter: Ruslan Dautkhanov
>Assignee: Shaofeng SHI
>Priority: Major
> Labels: cdh, cloudera, configuration, hive
>
> Getting following exception while trying to build a cube
> {noformat}
> java.lang.RuntimeException: Cannot find hive-site.xml in
> kylin_hadoop_conf_dir: /etc/hadoop/conf. In order to enable spark cubing, you
> must set kylin.env.hadoop-conf-dir to a dir which contains at least
> core-site.xml, hdfs-site.xml, hive-site.xml, mapred-site.xml, yarn-site.xml
> at
> org.apache.kylin.engine.spark.SparkExecutable.doWork(SparkExecutable.java:117)
> at
> org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:125)
> at
> org.apache.kylin.job.execution.DefaultChainedExecutable.doWork(DefaultChainedExecutable.java:64)
> at
> org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:125)
> at
> org.apache.kylin.job.impl.threadpool.DefaultScheduler$JobRunner.run(DefaultScheduler.java:144)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
> at java.lang.Thread.run(Thread.java:745)
> {noformat}
> I am using Kylin binaries for CDH downloaded from kylin.apache.org.
> Yes, indeed hive-site.xml is not in /etc/hadoop/conf in Cloudera's
> distribution for Hadoop.
> hive-site.xml is in /etc/hive/conf, not in /etc/hadoop/conf
> The other four files:
> core-site.xml, hdfs-site.xml, mapred-site.xml, yarn-site.xml
> can be found in /etc/hadoop/conf but, again, not hive-site.xml which is in
> /etc/hive/conf .
> Would be great to have this adjusted for CDH.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (KYLIN-3137) Update hive-site.xml locating logic for CDH distro
[
https://issues.apache.org/jira/browse/KYLIN-3137?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16497441#comment-16497441
]
ASF subversion and git services commented on KYLIN-3137:
Commit ad8d25f7c9646ad85395f5535fb28a97b8eb25bf in kylin's branch
refs/heads/KYLIN-3137 from shaofengshi
[ https://gitbox.apache.org/repos/asf?p=kylin.git;h=ad8d25f ]
KYLIN-3137 update kylin.properties for spark 2.1
> Update hive-site.xml locating logic for CDH distro
> --
>
> Key: KYLIN-3137
> URL: https://issues.apache.org/jira/browse/KYLIN-3137
> Project: Kylin
> Issue Type: Improvement
> Components: Job Engine, Others, Spark Engine
>Affects Versions: v2.2.0
>Reporter: Ruslan Dautkhanov
>Assignee: Shaofeng SHI
>Priority: Major
> Labels: cdh, cloudera, configuration, hive
>
> Getting following exception while trying to build a cube
> {noformat}
> java.lang.RuntimeException: Cannot find hive-site.xml in
> kylin_hadoop_conf_dir: /etc/hadoop/conf. In order to enable spark cubing, you
> must set kylin.env.hadoop-conf-dir to a dir which contains at least
> core-site.xml, hdfs-site.xml, hive-site.xml, mapred-site.xml, yarn-site.xml
> at
> org.apache.kylin.engine.spark.SparkExecutable.doWork(SparkExecutable.java:117)
> at
> org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:125)
> at
> org.apache.kylin.job.execution.DefaultChainedExecutable.doWork(DefaultChainedExecutable.java:64)
> at
> org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:125)
> at
> org.apache.kylin.job.impl.threadpool.DefaultScheduler$JobRunner.run(DefaultScheduler.java:144)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
> at java.lang.Thread.run(Thread.java:745)
> {noformat}
> I am using Kylin binaries for CDH downloaded from kylin.apache.org.
> Yes, indeed hive-site.xml is not in /etc/hadoop/conf in Cloudera's
> distribution for Hadoop.
> hive-site.xml is in /etc/hive/conf, not in /etc/hadoop/conf
> The other four files:
> core-site.xml, hdfs-site.xml, mapred-site.xml, yarn-site.xml
> can be found in /etc/hadoop/conf but, again, not hive-site.xml which is in
> /etc/hive/conf .
> Would be great to have this adjusted for CDH.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (KYLIN-3137) Update hive-site.xml locating logic for CDH distro
[
https://issues.apache.org/jira/browse/KYLIN-3137?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16496541#comment-16496541
]
ASF subversion and git services commented on KYLIN-3137:
Commit f11aec308119426222a4bc9d9d6a28f551a16111 in kylin's branch
refs/heads/KYLIN-3137 from shaofengshi
[ https://gitbox.apache.org/repos/asf?p=kylin.git;h=f11aec3 ]
KYLIN-3137 Spark cubing refine
> Update hive-site.xml locating logic for CDH distro
> --
>
> Key: KYLIN-3137
> URL: https://issues.apache.org/jira/browse/KYLIN-3137
> Project: Kylin
> Issue Type: Improvement
> Components: Job Engine, Others, Spark Engine
>Affects Versions: v2.2.0
>Reporter: Ruslan Dautkhanov
>Assignee: Shaofeng SHI
>Priority: Major
> Labels: cdh, cloudera, configuration, hive
>
> Getting following exception while trying to build a cube
> {noformat}
> java.lang.RuntimeException: Cannot find hive-site.xml in
> kylin_hadoop_conf_dir: /etc/hadoop/conf. In order to enable spark cubing, you
> must set kylin.env.hadoop-conf-dir to a dir which contains at least
> core-site.xml, hdfs-site.xml, hive-site.xml, mapred-site.xml, yarn-site.xml
> at
> org.apache.kylin.engine.spark.SparkExecutable.doWork(SparkExecutable.java:117)
> at
> org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:125)
> at
> org.apache.kylin.job.execution.DefaultChainedExecutable.doWork(DefaultChainedExecutable.java:64)
> at
> org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:125)
> at
> org.apache.kylin.job.impl.threadpool.DefaultScheduler$JobRunner.run(DefaultScheduler.java:144)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
> at java.lang.Thread.run(Thread.java:745)
> {noformat}
> I am using Kylin binaries for CDH downloaded from kylin.apache.org.
> Yes, indeed hive-site.xml is not in /etc/hadoop/conf in Cloudera's
> distribution for Hadoop.
> hive-site.xml is in /etc/hive/conf, not in /etc/hadoop/conf
> The other four files:
> core-site.xml, hdfs-site.xml, mapred-site.xml, yarn-site.xml
> can be found in /etc/hadoop/conf but, again, not hive-site.xml which is in
> /etc/hive/conf .
> Would be great to have this adjusted for CDH.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (KYLIN-3137) Update hive-site.xml locating logic for CDH distro
[
https://issues.apache.org/jira/browse/KYLIN-3137?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16496282#comment-16496282
]
Shaofeng SHI commented on KYLIN-3137:
-
I tested several ways to specify the hive-site.xml, but none of them work:
* export HIVE_CONF_DIR=/etc/hive/conf && spark-submit ...
* --conf spark.yarn.dist.files=path/to/your/hive-site.xml
* --files path/to/your/hive-site.xml
Only two way can work:
* Put hive-site.xml to HADOOP_CONF_DIR (as-is way)
* Put hive-site.xml to spark/conf
> Update hive-site.xml locating logic for CDH distro
> --
>
> Key: KYLIN-3137
> URL: https://issues.apache.org/jira/browse/KYLIN-3137
> Project: Kylin
> Issue Type: Improvement
> Components: Job Engine, Others, Spark Engine
>Affects Versions: v2.2.0
>Reporter: Ruslan Dautkhanov
>Assignee: Shaofeng SHI
>Priority: Major
> Labels: cdh, cloudera, configuration, hive
>
> Getting following exception while trying to build a cube
> {noformat}
> java.lang.RuntimeException: Cannot find hive-site.xml in
> kylin_hadoop_conf_dir: /etc/hadoop/conf. In order to enable spark cubing, you
> must set kylin.env.hadoop-conf-dir to a dir which contains at least
> core-site.xml, hdfs-site.xml, hive-site.xml, mapred-site.xml, yarn-site.xml
> at
> org.apache.kylin.engine.spark.SparkExecutable.doWork(SparkExecutable.java:117)
> at
> org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:125)
> at
> org.apache.kylin.job.execution.DefaultChainedExecutable.doWork(DefaultChainedExecutable.java:64)
> at
> org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:125)
> at
> org.apache.kylin.job.impl.threadpool.DefaultScheduler$JobRunner.run(DefaultScheduler.java:144)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
> at java.lang.Thread.run(Thread.java:745)
> {noformat}
> I am using Kylin binaries for CDH downloaded from kylin.apache.org.
> Yes, indeed hive-site.xml is not in /etc/hadoop/conf in Cloudera's
> distribution for Hadoop.
> hive-site.xml is in /etc/hive/conf, not in /etc/hadoop/conf
> The other four files:
> core-site.xml, hdfs-site.xml, mapred-site.xml, yarn-site.xml
> can be found in /etc/hadoop/conf but, again, not hive-site.xml which is in
> /etc/hive/conf .
> Would be great to have this adjusted for CDH.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (KYLIN-3137) Update hive-site.xml locating logic for CDH distro
[
https://issues.apache.org/jira/browse/KYLIN-3137?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16306400#comment-16306400
]
Ruslan Dautkhanov commented on KYLIN-3137:
--
Yep that's how I worked around this - created a symlink.
Although CM may recreate /etc/hadoop/conf as this directory is managed by
Cloudera Manager and my symlink will be gone.
I was thinking that Kylin should also autodetect and set HIVE_CONF_DIR
(defaults to /etc/hive/conf for CDH), not just HADOOP_CONF_DIR..
> Update hive-site.xml locating logic for CDH distro
> --
>
> Key: KYLIN-3137
> URL: https://issues.apache.org/jira/browse/KYLIN-3137
> Project: Kylin
> Issue Type: Bug
> Components: General
>Affects Versions: v2.2.0
>Reporter: Ruslan Dautkhanov
> Labels: cdh, cloudera, configuration, hive
>
> Getting following exception while trying to build a cube
> {noformat}
> java.lang.RuntimeException: Cannot find hive-site.xml in
> kylin_hadoop_conf_dir: /etc/hadoop/conf. In order to enable spark cubing, you
> must set kylin.env.hadoop-conf-dir to a dir which contains at least
> core-site.xml, hdfs-site.xml, hive-site.xml, mapred-site.xml, yarn-site.xml
> at
> org.apache.kylin.engine.spark.SparkExecutable.doWork(SparkExecutable.java:117)
> at
> org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:125)
> at
> org.apache.kylin.job.execution.DefaultChainedExecutable.doWork(DefaultChainedExecutable.java:64)
> at
> org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:125)
> at
> org.apache.kylin.job.impl.threadpool.DefaultScheduler$JobRunner.run(DefaultScheduler.java:144)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
> at java.lang.Thread.run(Thread.java:745)
> {noformat}
> I am using Kylin binaries for CDH downloaded from kylin.apache.org.
> Yes, indeed hive-site.xml is not in /etc/hadoop/conf in Cloudera's
> distribution for Hadoop.
> hive-site.xml is in /etc/hive/conf, not in /etc/hadoop/conf
> The other four files:
> core-site.xml, hdfs-site.xml, mapred-site.xml, yarn-site.xml
> can be found in /etc/hadoop/conf but, again, not hive-site.xml which is in
> /etc/hive/conf .
> Would be great to have this adjusted for CDH.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
[jira] [Commented] (KYLIN-3137) Update hive-site.xml locating logic for CDH distro
[
https://issues.apache.org/jira/browse/KYLIN-3137?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16305928#comment-16305928
]
Shaofeng SHI commented on KYLIN-3137:
-
This is normal (not a problem of CDH). When submit Spark, only one
HADOOP_CONF_DIR can be set. To be simple, just copy hive-site.xml to or create
a simbolic link for it in that folder.
> Update hive-site.xml locating logic for CDH distro
> --
>
> Key: KYLIN-3137
> URL: https://issues.apache.org/jira/browse/KYLIN-3137
> Project: Kylin
> Issue Type: Bug
> Components: General
>Affects Versions: v2.2.0
>Reporter: Ruslan Dautkhanov
> Labels: cdh, cloudera, configuration, hive
>
> Getting following exception while trying to build a cube
> {noformat}
> java.lang.RuntimeException: Cannot find hive-site.xml in
> kylin_hadoop_conf_dir: /etc/hadoop/conf. In order to enable spark cubing, you
> must set kylin.env.hadoop-conf-dir to a dir which contains at least
> core-site.xml, hdfs-site.xml, hive-site.xml, mapred-site.xml, yarn-site.xml
> at
> org.apache.kylin.engine.spark.SparkExecutable.doWork(SparkExecutable.java:117)
> at
> org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:125)
> at
> org.apache.kylin.job.execution.DefaultChainedExecutable.doWork(DefaultChainedExecutable.java:64)
> at
> org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:125)
> at
> org.apache.kylin.job.impl.threadpool.DefaultScheduler$JobRunner.run(DefaultScheduler.java:144)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
> at java.lang.Thread.run(Thread.java:745)
> {noformat}
> I am using Kylin binaries for CDH downloaded from kylin.apache.org.
> Yes, indeed hive-site.xml is not in /etc/hadoop/conf in Cloudera's
> distribution for Hadoop.
> hive-site.xml is in /etc/hive/conf, not in /etc/hadoop/conf
> The other four files:
> core-site.xml, hdfs-site.xml, mapred-site.xml, yarn-site.xml
> can be found in /etc/hadoop/conf but, again, not hive-site.xml which is in
> /etc/hive/conf .
> Would be great to have this adjusted for CDH.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
